University of Southern California Dissertations and Theses

Writing about worries as an intervention for test anxiety in undergraduates

(USC Thesis Other)

Writing about worries as an intervention for test anxiety in undergraduates

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: WRITING ABOUT WORRIES 1

WRITING ABOUT WORRIES AS AN INTERVENTION FOR

TEST ANXIETY IN UNDERGRADUATES

by

Nicholas Blank-Spadoni

A Dissertation Presented to the
FACULTY OF THE USC ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION

August 2013

Copyright 2013 Nicholas Blank-Spadoni
WRITING ABOUT WORRIES 2

Acknowledgements

I would like to thank my committee: Drs. Kimberly Hirabayashi, Helena Seli, Bob Keim,
and Mary-Helen Immordino-Yang. The project never would have gotten off the ground without
the early guidance of Dr. Hirabayashi, and Dr. Seli’s quick action and guidance at times of
perceived crisis helped keep the project alive while I panicked over whether or not there was still
a pulse to be found. Dr. Keim’s ability to open locked doors—sometimes literally—both before
and after data collection was a priceless asset for me. Finally, Dr. Immordino-Yang’s work that I
stumbled across long before I made the decision to become a Trojan, and subsequently her
sending me a paper, under review at the time, about the brain at rest, all became the clay that I
would sculpt into this dissertation. For that, I am profoundly grateful.

In addition to assistance from my committee, data collection would not have been
possible without the assistance of Diane Melrose to whom I am very grateful for providing
administrative assistance and access to her students. The support of Nancy Guirguis in the final
weeks of this project was instrumental in its completion and I could not possibly express the
value of her assistance: a million thanks are not enough.

This dissertation owes its existence to the help of Tony Le and Jock Milspaugh. They
say friends help friends move. They have no idea.

I owe a huge debt of gratitude to Krysten Lewis Nunn, the best person a person could
ask for. A dissertation would be the least of my problems if you weren’t there to remind me just
what can be done when we give ourselves permission to go, and to do, and to be.

I would also like to thank Leonard Spadoni for his informal contributions to both my
developing this project before there was a project to be developed and to maintaining my sanity
once the project was under development.
WRITING ABOUT WORRIES 3
Table of Contents
List of Tables 5
Abstract 6
Chapter One: Overview of the Study 7
Introduction 7
Background 8
Statement of the Problem 10
Purpose of the Study 10
Research Questions and Hypotheses 11
Significance of the Study 11
Methodology 11
Sample 1 12
Sample 2 13
Assumptions 13
Definition of Terms 14
Organization of the Study 17
Chapter Two: Literature Review 19
Anxiety 19
Distinguishing Stress and Anxiety 19
Anxiety Disorders 20
General Anxiety Disorder and Social Anxiety Disorder 20
State and Trait Anxiety 22
Subclinical Anxiety 22
Models of Test Anxiety 23
The Cognitive Turn 27
Information Processing Theory 27
Processing Efficiency Theory and Attentional Control Theory 29
Neural Correlates of Attentional Control 32
Implications for Interventions 34
Educational Implications of Attentional Control and DMN 34
Implications in Health Sciences Education 34
Recent Interventions Among Other Populations 39
Conclusion 42
Chapter Three: Methodology 44
Research Questions 44
Research Design 44
Population and Sample 46
Instrumentation 47
Demographics Questionnaire 47
Cognitive Test Anxiety Scale 47
Trait Anxiety Scale 48
State Anxiety Scale 48
Intervention Group and Treatment Group Writing Prompts 48
Data Collection 49
Sample 1 48
WRITING ABOUT WORRIES 4
Sample 2 50
Quantitative Analysis 50
Analysis of Writing Samples 51
Chapter Four: Results 52
Intercorrelations 52
Research Question 1: Exam Outcomes 54
Research Question 2: State Anxiety 54
Analysis of Writing Samples 56
Chapter Five: Discussion 57
Hypotheses about Performance and State Anxiety 57
Processing Efficiency Versus Processing Effectiveness 59
Word Count as Indicator of Fidelity to Treatment 60
Preintervention Level of State Anxiety 62
Level of Cognitive Test Anxiety 64
Explicit Instruction in Compensatory Strategies 65
Limitations of the Study 69
Delimitations of the Study 67
Implications for Students, Instructors, Support Personnel, and Administration 68
Recommendations for Further Research 69
Recommendations for Research in Test Anxiety 69
Recommendations for Research in Attentional Control and DMN 71
Conclusion 73
References 75
Appendix 1: Demographics Questionnaire 90
Appendix 2: Cognitive Test Anxiety Scale 91
Appendix 3: Trait Anxiety Scale 95
Appendix 4: State Anxiety Scale 97
Appendix 5: Prompt for Treatment Group 98
Appendix 6: Prompt for Control Group 99
WRITING ABOUT WORRIES 5
List of Tables
Table 1: Demographic Information for Sample 1 and Sample 2 47
Table 2: Timeline of data collection across samples 50
Table 3: Intercorrelations between variables for Sample 1 53
Table 4: Intercorrelations between variables for Sample 2 54
Table 5: Intercorrelations between variables across Sample 1 and Sample 2 55
WRITING ABOUT WORRIES 6

Abstract

The phenomenon of test anxiety has long been associated with decrements in
performance (Zeidner, 1998) and has been found to affect up to 40% of all students (Cizek &
Berg, 2006). Because a substantial number of students that deal with test anxiety perform below
their ability on exams, test anxiety during cognitive ability tests has also been identified as a root
cause of differential predictive validity of academic performance (Bonnaccio, Reeve & Winford,
2011). In addition, recent developments in cognitive psychology and neuroscience have led to a
greater understanding of the neurological and psychological mechanisms at work in test anxiety
(e.g., Eysenck & Derakshan, 2011; Immordino-Yang, Christodolou & Singh, 2012). Together,
the well-known decrements in performance attributable to test anxiety, the hardships students
suffer as a result of these decrements, and the increasing influence of test performance on
decision-making in a number of educational settings contribute to a situation where developing
effective test anxiety interventions is of the utmost theoretical, ethical, and practical importance.

Given the importance of developing effective interventions for test anxiety, there is,
relative to the rather large body of literature that looks at the symptoms, causes, and effects of
test anxiety, very little research that offers empirical findings from studies that explore effective
interventions for test anxiety. The current research sought to replicate findings from a recent
study that found a 10-minute expressive writing intervention immediately before a final exam
allowed test anxious students to overcome their anxieties and outperform their less-anxious peers
(Ramirez & Beilock, 2011). Students in the Ostrow School of Dentistry and Rossier School of
Education at the University of Southern California performed the same kind of expressive
writing intervention during a final exam. Survey data regarding cognitive test anxiety levels and
trait anxiety were collected at Time 1 and the intervention was performed during at Time 2,
during which levels of state anxiety were be assessed immediately before and after the
intervention to determine the effect of the intervention on state anxiety.

Ultimately, the current research sought to answer the following questions: 1) Is there a
difference in exam outcomes for students who write about worries immediately before an exam
and for students who do not? 2) Does writing about worries immediately before an exam reduce
self-report survey scores for state anxiety? Findings showed that condition differences did not
exert a significant effect on exam performance when controlling for prior exam score in Sample
1 or Sample 2, but word count produced during the intervention significantly predicted exam
performance across both samples when controlling for prior exam score. Further, the
intervention significantly reduced state anxiety in Sample 1, but not Sample 2.

Recommendations for future research include additional investigation amongst samples
within a more homogenous context; qualitative analysis of the content of writing performed as
part of the intervention; comparative investigation of similar interventions that induce
constructive internal reflection, mind-wandering, and interoception; and consideration of
research designs that might allow for examination of the effects DMN deactivation induction in
educational contexts.
WRITING ABOUT WORRIES 7

CHAPTER ONE: OVERVIEW OF THE STUDY

Introduction

It is widely accepted that test anxiety has negative effects on cognitive performance. Test
anxiety has been found to have negative effects on the storage and processing of information in
working memory (Dutke & Stuber, 2001; Ikeda, Iwanga, & Seiwa, 1996; Mueller, 1980; Naveh-
Benjamin, 1991), academic performance (Biedel & Turner, 1988; Cassady & Johnson, 2002;
Chapelle et al, 2005; Morris, Davis, & Hutchings, 1981; Schutz & Davis, 2000; Sub & Prabha,
2003), standardized achievement tests, (Everson, Millsap, & Rodriguez, 1991), retention in
academic programs (Tobias, 1979), and differential predictive validity of cognitive ability and
academic performance (Bonaccio, Reeve, & Winford, 2012; Reeve & Bonaccio, 2008).

While models that seek to explain the mechanism by which test anxiety exerts its
influence on performance have changed in past the 60 years over which the construct has been
closely examined, the definition of test anxiety has remained largely unchanged. Sarason and

Mandler (1952) defined test anxiety as “feelings of inadequacy, helplessness, heightened somatic
reaction, anticipations of punishment, or loss of status and esteem, and implicit, attempts at
leaving the test situation” (p. 166). The prevalence of this phenomenon is often reported in the
range of 25 to 30% of all students (Hill & Wigfield, 1984; McGuire, Mitic & Neumann, 1987)
with higher reports reaching upward of 40% (Cizek & Berg, 2006). It has been argued that as it
becomes increasingly common for educational administrators and policy makers to rely on test
data in decision making processes, especially at the K-12 level but also in higher education, the
stakes in educational testing will continue to rise and so too will the prevalence of test anxiety
(Casbaro, 2005).
WRITING ABOUT WORRIES 8

Given that test anxiety is known to exert a negative influence on performance in
evaluative situations for such a significant number of students, it is of the utmost ethical,
theoretical, and practical importance to better understand the ways in which this influence is
exerted on cognitive ability tests. Cognitive ability tests include those that measure IQ,
scholastic aptitude, or fitness for admission to undergraduate, graduate, or professional
programs of study as well as academic progress within programs of study. Data from these
cognitive ability tests are often misinterpreted without proper understanding and consideration
of performance decrements associated with test anxiety, which leads to a situation where the
validity of the whole enterprise of testing is challenged (Hembree, 1988).

Background of the Problem

It is widely accepted that performance in evaluative situations has far reaching
implications for students, educators, administrators, and educational policy makers. For
students, performance in evaluation situations often determines what academic programs a
student will have access to. In cases where it is applicable, the amount of financial support a
student will have access to in those programs is often a matter of performance in evaluative
situations. In addition, for educational or professional endeavors that culminate in certification
or licensure, one’s ultimate ability to participate in professional or academic field may rely on
performance in evaluative situations. One setting in particular where examinations traditionally
make up the bulk of educational assessment and where stress and anxiety has been found at
high levels is that of health science education.

High volumes of stress and anxiety are well documented among students of health sciences
(e.g., Beggs, Shileds, & Janiszewski Goodin, 2011; Graham, Vettarino, Seifeldin, Singal, 2010; Ping,
Subramaniam & Krishnaswamy, 2008; Powel, 2004; Sanisgiry & Sail, 2005).
WRITING ABOUT WORRIES 9

While anxiety is amongst the most studied of psychological phenomena, a great deal of that
research deals with identifying the causes, symptoms, and effects of anxiety, primarily through
surveys and interviews. Relative to this type of research, there is very little research that focuses
specifically on test anxiety, takes place in classroom settings, and provides empirical data about
developing useful interventions for test anxiety. The paucity of research into effective
interventions of test anxiety in the setting of health sciences education is especially relevant in
light of the burgeoning body of literature that offers explanatory models for understanding the
psychological and biological mechanisms by which test anxiety exerts influence on performance
in evaluative situations. One pattern clearly visible in this literature is that it is only very rarely
the case that an explicit attempt is made to translate biological and psychological explanations
of test anxiety into applicable educational interventions. This is especially relevant given
findings that could potentially necessitate a reconsideration of the relation between the most
fundamental aspects of test anxiety—the cognitive and the affective.

Traditional inquiry into mechanisms and effects of test anxiety has considered cognitive and
affective aspects of test anxiety largely in isolation. While there is evidence that the cognitive aspect
of test anxiety exerts a much greater influence on decrements in performance than the affective
aspect, recent research has brought this decades-old assumption into question. Recent research
yields findings that suggest a more nuanced relationship between cognitive and affective aspects of
performance in evaluative situations. This relationship can only be understood at the deepest level
through explanatory models that address both cognitive and affective aspects of the phenomena
(Mattarella-Micke, Mateo, Kozak, Foster, & Beilock, 2011).

One research paradigm that has lent itself well to the debate over this reconsideration of
the interplay between cognitive and affective aspects of test anxiety is Pennebaker’s expressive
WRITING ABOUT WORRIES 10

writing paradigm (Pennebaker, 1994). This paradigm holds that writing about a past traumatic
event or about an impending anxiety-inducing event reduces worry, increases working memory,
and could lead to improvements in performance (cf. Smyth, 1999; Klien & Boals, 2001;
Ramirez & Beilock, 2011). It has been argued that thinking or writing about negative
information makes it more salient in memory rather than less salient (Lang & Lang, 2011;
Schmader & Johns, 2003; Wegner, Schneider, Carter, & Wite, 1987). Additionally,
Pennebaker’s expressive writing paradigm proposes treatments that entail performing writings
about emotions over an extended period of time (Klein & Boals, 2001). However, recent work
has shown that writing about worries concerning performance immediately before performance
provides significant increases in performance (Ramirez & Beilock, 2011).

Statement of the Problem

There is an abundant literature that identifies varied causes, symptoms, and negative
effects of test anxiety. A subset of this literature indicates an issue with high levels of anxiety in
health sciences education and calls for the development of effective interventions. There appears
to be, however, relatively little work that centers on effective interventions for test anxiety in
health sciences education.

Purpose of the Study

The purpose of the current research is to extend text anxiety literature beyond identifying
causes, symptoms, and effects of test anxiety by providing empirical data from a test anxiety
intervention and seeking to integrate insights from lines of research dealing with attentional
control and the default mode network (DMN). More specifically, the purpose of the current
study is to see if writing about worries regarding the outcome of an exam immediately before the
exam reduces levels of state anxiety and improves performance.
WRITING ABOUT WORRIES 11

Research Questions and Hypotheses

The current study centers on two questions:

1. Is there a difference in exam outcomes for students who write about worries
immediately before an exam and for students who do not?

2. Does writing about worries immediately before an exam reduce self-report survey scores
for state anxiety?

The following were hypotheses regarding the research questions:

1. Writing about worries will increase test performance for students.

2. The intervention will reduce state anxiety for high and low test anxiety students, with
a greater reduction for students with greater test anxiety.

Significance of the Study

This study has implications for confirming the recent work of Ramirez and Beilock
(2011), who showed that a 10-minute writing intervention enabled students with high levels of
text anxiety to overcome their anxieties and outperform students with lower levels of test anxiety
to the extent that the findings have been described as “almost too good to be true” (Geller, 2011,
n.p.). An intervention that enables students with high levels of stress to overcome their anxieties
has immediate implications in the treatment of test anxiety and offers potential support for
theoretical explanations of the mechanisms by which test anxiety affects performance. The
nature of the intervention is also uniquely placed to speak to a need to reconsider the relationship
between cognitive and affective processes, especially given emerging psychobiological research
into the DMN and its functional correlation to the task-positive network in brain.

Methodology
WRITING ABOUT WORRIES 12

Sample 1. Survey data regarding participant’s demographic information, trait anxiety,
and cognitive test anxiety were gathered at Time 1 after an in-class introduction to the study held
twice: once for 1
st
-year dental hygiene students and a second time for 2
nd
-year dental hygiene
students. Participation was explained to be voluntary and uncompensated. Subsequent to
completing one exam with no intervention, which was used to control for prior exam score,
survey data were collected identifying state anxiety levels at Time 2, immediately before the
second of eight tests for 2
nd
-year students and immediately before the final for 1
st
year students.
Immediately following collection of state anxiety levels, students performed the intervention
consisting of writing for 10 minutes about their worries about the test (treatment) or thinking
about what they thought would not be covered on the test (control). Immediately following the
intervention, students completed the same state anxiety survey. At that point, students began the
test.

To answer the first research question regarding the interaction between intervention
and exam performance, I used ANCOVA analysis to examine the effect of condition
differences on exam performance, controlling for prior exam performance. Subsequently,
regression analysis was used to examine if total word count of the writing produced during the
intervention predicted performance.

To answer the second research question regarding the interaction between the
intervention and state anxiety levels, T-tests were computed to determine if mean scores of
preintervention state anxiety differed from postintervention state anxiety. Subsequently,
regression analysis was used to examine if the level of preintervention state anxiety significantly
predicted the amount of change between preintervention state anxiety and postintervention state
anxiety. Finally, regression analysis was used to examine if the level of cognitive test anxiety
WRITING ABOUT WORRIES 13

significantly predicted the amount of change between preintervention state anxiety
and postintervention state anxiety.

Sample 2. Survey data regarding participants’ demographic information, trait anxiety,
and cognitive test anxiety were gathered at Time 1 with Qualtrics after the study was introduced
to an undergraduate learning skills and motivation course. Extra credit was offered for
participation. The study was introduced subsequent to administration of the midterm exam,
which was used to control for prior exam scores. Survey data were collected identifying state
anxiety levels at Time 2, immediately before the final exam. Immediately following collection
of state anxiety levels, students performed the intervention consisting of writing for 10 minutes
about their worries about the test (treatment) or thinking about what they thought would not be
covered on the test (control). Immediately following the intervention, students completed the
same state anxiety survey. At that point, student began the test. The same analyses were used to
answer the same questions as in Sample 1.

Assumptions

This study operated under a series of assumptions. First, this study assumed that self-
report survey measures are valid, reliable, and provide an accurate account of the construct they
measure. There has been some debate about the use of self-report survey data but the author
assumed that work confirming the reliability and validity of the measures used provided a
sufficient foundation on which to base the study. The second assumption this study made is that
testing situations in medical education induce varying levels of test anxiety across students
taking the test.

Another assumption is that writing about worries before the test elicits a certain mechanism,
described in the literature review, that intervenes against decrements in performance
WRITING ABOUT WORRIES 14

attributable to test anxiety. As such, the design of the study also operated under the assumption
that the content of the thought represented in writing performed for the intervention determines
the elicitation of this mechanism and therefore assumes that the sham treatment condition of
having students think about what they expect to not be on the exam will not elicit this
mechanism.

Definition of Terms

Attentional Control Theory

A theory developed by Eyesenck, Derakshan, Santos, & Calvo (2007) that extends
Performance Efficiency Theory (Eysenck & Calvo, 1992) primarily by addressing imprecise
explanations of the functions of central executive system of working memory and the
mechanisms by which test anxiety effects them. The theory holds that performance decrements
attributable to test anxiety are mediated by attentional control, in particular the inhibiting and
shifting function of the central executive.

Central Executive

A modality-free subcomponent of Working Memory responsible for coordinating the
phonological loop and visuospatial sketchpad (Baddeley & Hitch, 1974; Baddeley 1986;
2000; 2010).

Default Mode Network (DMN)

A network of neurobiological activity implicated in cognitively undemanding tasks and
resting mental states that plays a role in internally directed, stimulus-independent processes
(Immordino-Yang, Christodoulou, & Singh, 2012) like environmental or visuospatial
monitoring, social and self-referential processing, emotion processing and emotional memory,
and autobiographical memory (Drevets, 2001; Fales, et al., 2008; Grecius, Krasnow, Reiss &
WRITING ABOUT WORRIES 15

Menon, 2003; Gusnard & Raichele, 2001; Maddock, 1999; Maddock, Garrett, & Buonocore,
2003; Mazoyer et al., 2001).

Emotionality

The affective aspect of test anxiety, generally only correlated with performance
decrements in the presence of high levels of worry (Deffenbacher, 1980; Morris et al., 1981;
Schwarzer 1984), characterized by heightened somatic activity including dizziness, sweating,
nausea or nervous stomach, increased heart rate, and shallow or rapid breathing when
encountering an evaluative situation (Sarason & Mandler, 1954; Liebert & Morris, 1967).

Expressive Writing

A paradigm of psychological treatment characterized by the belief in increased physical
and mental benefits afforded by writing that entails a component self-disclosure of thoughts and
feelings (Klein & Boals, 2001).

Inhibition

A specific function of the central executive that deals with the suppression of task-
irrelevant stimuli and responses (Miyake et al., 2000).

Interference Models of Test Anxiety

An explanatory model of test anxiety that holds test anxiety is the cause of performance
decrements on account task-irrelevant stimuli, responses, or thoughts that interfere with task-
relevant stimuli, responses, or thoughts (Wine, 1971; Beck, Emery, & Greenberg, 1985).

Phonological Loop

A subcomponent of Working Memory responsible for short-term transient storage and
processing of verbal and acoustic information (Baddeley & Hitch, 1974; Baddeley 1986;
2000; 2010).
WRITING ABOUT WORRIES 16

Shifting

A specific function of the central executive that deals with switching attention from one
task to another task (Miyake et al., 2000).

State anxiety

A situational, transient, acute onset form of anxiety (Spielberger, 1983;
Spielberger, Gaudry, & Vagg, 1975; Spielberger, Gorsuch, & Lushene,1970).

Stress

A relationship between the person and the environment that is appraised by the person
as taxing or exceeding his or her resources and endangering his or her well-being (Lazarus &
Folkman, 1984, p. 21).

Task-Positive Network

A network of neurobiological activity implicated in actively engaging outward-
looking, goal-directed, externally directed attention-demanding tasks involved in cognitive
control (Immordino-Yang, Christodoulou, & Singh, 2012).

Test Anxiety

Feelings of inadequacy, helplessness, heightened somatic reaction, anticipations of
punishment, or loss of status and esteem, and implicit attempts at leaving the test situation in
response to evaluative situations (Sarason & Mandler, 1952, P. 166).

Trait anxiety

A stable, characteristic-like, chronic propensity to encounter state anxiety (Spielberger,
1983; Spielberger, Gaudry, & Vagg, 1975; Spielberger, Gorsuch, & Lushene,1970).

Visuospatial Sketchpad
WRITING ABOUT WORRIES 17

A subcomponent of Working Memory responsible for short-term transient storage and
processing of visual and spatial information (Baddeley & Hitch, 1974; Baddeley 1986; 2000;
2010).

Working Memory

Based on Baddeley’s model (Baddeley & Hitch, 1974; Baddeley 1986; 2000; 2010) one
component, consisting of several subcomponents, of Information Processing Theory
responsible for short-term transient storage and processing of information.

Worry

The cognitive aspect of test anxiety characterized by expectations of aversive
consequences that become activated, especially in high trait anxious individuals, in stressful
situations that involve evaluative and/or competitive components and which includes forms of
self-dialogue and verballinguistic mentation (Borkovec, 1994).

Organization of the Study

Thus far, the study has provided an introduction to the problem; background of the
problem; statement of the problem; the purpose of the study; the research question and
hypotheses; the significance of the study; a brief description of the methodology; discussion of
assumptions, limitations, and delimitations; and definition of relevant terms. Chapter 2 proceeds
with a review of the literature relevant to a discussion of anxiety; early conceptions of
explanatory models of test anxiety; Information Processing Theory and working memory;
Attentional Control Theory, DMN, and implications for interventions. Chapter 3 identifies
methodology used in the study including a detailed description of the research design and
procedures, the population and sampling procedure, and instruments and their selection as well
as the validity and reliability of instruments. Chapter 4 presents the data collected, data analyses,
WRITING ABOUT WORRIES 18

and results. Chapter 5 discusses the results and culminates in conclusions as well as provides

recommendations for future research.
WRITING ABOUT WORRIES 19

CHAPTER TWO: LITERATURE REVIEW

Anxiety

This chapter is divided into four sections. The first section explores the construct of
anxiety by differentiating stress from anxiety; discusses clinical anxiety disorders, especially
Generalized Anxiety Disorder and Social Anxiety Disorder; defines the difference between state
and trait anxiety; and closes with a discussion of the relevance of subclinical anxiety. The
second section provides an overview of the historical trajectory of explanatory models of test
anxiety through Sarason and Mandler’s (1954) influence on Liebert and Morris (1967) and

Liebert and Morris’ subsequent influence on Wine (1971). The emergence of skills deficit
models of test anxiety popular in the 1980s is also discussed. The third section examines
currently dominant explanatory models of test anxiety by reconstructing the development of
Processing Efficiency Theory and its evolution into Attentional Control Theory and then
concludes with a discussion of the insights from the neural correlates of attentional control.
Finally, the fourth section frames the implications for education of the burgeoning research
that supports Attentional Control Theory, including support from recent research on the DMN,
and discusses the state of research that offers empirical findings about test anxiety
interventions in the health sciences education setting as well as in other settings.

Distinguishing Stress and Anxiety

The transactional model of stress and emotion is one of the more influential models of
psychological stress. This model views stress as a dynamic transaction between a person and the
environment during which a person is threatened or challenged by his or her environment and
must appraise and respond to the threat or challenge (Gatchel, Baum, & Krantz, 1989; Lazarus &
Folkman, 1984). Anxiety, a basic emotion, signals the cognitive appraisal of threat or
WRITING ABOUT WORRIES 20

uncertainty in the environment and has been found to be one of the most important and
pervasive reactions to stress (Sarason, 1986). In short, when stress from environmental stressors
continues unabated by resolution or adaptive coping, the fear-like response that might result is
what is generally referred to as anxiety. Because anxiety exerts such a significant influence on
cognitive, affective, and behavioral aspects of life (Zeidner, 1988), it is helpful to understand
categorization schemes for different types of anxiety in approaching an understanding of the
causes, symptoms, effects, and interventions for test anxiety.

Anxiety Disorders

Anxiety can be defined in large part by symptoms of excessive worry and autonomic
physiological arousal (Smith, Elwood, Feldner, & Olatunji, 2010). In a clinical context, there are
five recognized anxiety disorders for adults and a sixth, separation anxiety, that includes a
childhood onset (Klein, 2009). The five adult onset anxiety disorders recognized by the fourth
and most current edition of the Diagnostic and Statistical Manual are panic disorder,
agoraphobia, specific phobia, generalized anxiety disorder, and social anxiety (APA, 1994).
However, any treatment of separation anxiety, panic disorder, agoraphobia, and specific phobia
will be beyond the scope of the present study. Even so, it will still be instructive to look more
closely at General Anxiety Disorder and Social Phobia Disorder in to understand the relationship
between trait and state anxiety, measurements which this study employs in its analyses.

General Anxiety Disorder and Social Anxiety Disorder

More directly related to the present study are general anxiety disorder (GAD) and social
anxiety disorder (SAD). GAD is characterized by a consistent pattern of worry with no specific
source or content (Klein, 2009). That is, GAD refers to a consistently high level of anxiety that
is not necessarily related to a situational onset. A diagnosis of GAD consists of meeting certain
WRITING ABOUT WORRIES 21

diagnostic criteria. Those criteria are (a) at least six months of excessive anxiety and worry, (b)
significant difficulty in controlling the anxiety, (c) the presence for most days over the previous
six months of three or more of the following symptoms: (i) feeling wound-up, tense, or
restless, (ii) easily becoming fatigued or worn out, (iii) concentration problems, (iv) irritability,
(v) significant tension in muscles, and (iv) difficulty with sleep, (d) the symptoms are not part
of another disorder, (e) the symptoms cause clinically significant distress, and (f) the condition
is not due to a substance or medical issue (APA, 1994).

SAD is characterized by performance and interpersonal anxiety and can be specified as
generalized if anxiety occurs in most social situations (Klein, 2009). The diagnostic criteria of
SAD consist of (a) persistent and intense fear of social or performance situations, (b) exposure to
social situations resulting in an immediate intense anxiety reaction potentially taking the form of
a situationally bound or situationally predisposed panic attack , (c) recognition of the fear or
anxiety experienced as unreasonable, (d) marked avoidance of social situations when possible
and endured with dread when avoidance is not possible, (e) the symptoms cause clinically
significant distress, (f) symptoms persist for longer than 6 months, (g) the symptoms are not due
to a substance or medical issue, and (h) the symptoms are not attributable to another disorder
(APA, 1994).

Looking at the definitions and diagnostic criteria for GAD and SAD shows that there are
certainly some similarities between the two. Those similarities include difficulty in managing
the onset or extinction of the manifestation of persistent symptoms recognized as unreasonable
not due to other substance, medical, or psychological conditions. The presence of these
similarities indicates that is conceptual overlap among anxiety disorders. Indeed, there has been
and remains a great deal of debate about how to classify and diagnose anxiety disorders (cf.
WRITING ABOUT WORRIES 22

APA, 1980; APA, 1984; APA, 2013; Feldman, 1993; Gotlib, 1984). However, GAD and SAD
differ in a number of very important ways. The most noticeable is the generalized, non-
situational nature of GAD symptoms and the situation specific onset of SAD. The difference
between these two types of anxiety—a generalized, non-situational type on the one hand and an
acute onset situation-specific type on the other—is precisely the difference that defines state and
trait anxiety.

State and Trait Anxiety

Among the most important differences between GAD and SAD is that GAD is
characterized by an ongoing sense of anxiety with no particular situational catalyst while social
phobia is characterized by an acute, situation-specific anxiety response. These two different
kinds of anxiety responses—state and trait—are fundamental in understanding effects of and
treatment for different types of anxiety and will be especially important in understanding
different aspects of test anxiety. Work by Spielberger (e.g., Spielberger, 1983; Spielberger,
Gaudry, & Vagg, 1975; Spielberger, Gorsuch, & Lushene,1970) has been instrumental in
institutionalizing the distinction between state and trait anxiety. In addition, Spielberger (1979)
has developed an entire state-trait personality inventory which considers both state and trait
versions of anger as well as depression. The state-trait distinction is important because both state
anxiety and trait anxiety have fine-grained implications for both clinical and subclinical levels of
anxiety. We will turn now to a discussion of subclinical anxiety, into which most instantiations
of test anxiety would be categorized.

Subclinical Anxiety

The defining characteristics of GAD and SAD are important because some of the shared
characteristics and some of their unique characteristics are implicated in issues in a non-clinical
WRITING ABOUT WORRIES 23

population. Even though some people might not meet the diagnostic criteria for GAD or SAD,
many still suffer from some or all of the criteria they share but might not be affected to the extent
that would entail a clinical diagnosis or might not meet all of the criteria a clinical diagnosis
would entail. One such subclinical area in anxiety centers on the construct of test anxiety.

In summary, to understand test anxiety, a very specific form of anxiety, it is important to
understand the larger organizational framework of clinical anxiety disorders. While a thorough
discussion of each type of anxiety disorder is beyond the scope of this study, it is important to
note some of defining characteristics of GAD and SAD—namely the chronic generalized non-
situational nature of GAD and the acute onset of situationally induced anxiety that characterizes
SAD. This particular difference also characterizes the difference between state and trait anxiety.
Understanding the difference between state and trait anxiety are fundamental to understanding
instantiations of test anxiety.

Models of Test Anxiety

Sarason and Mandler’s (1952) Test Anxiety Theory which viewed test anxiety as
primarily a learned motivational response. Students who had experiences with tests that

“manifested as feelings of inadequacy, helplessness, heightened somatic reaction, anticipations
of punishment, or loss of status and esteem, and implicit attempts at leaving the test situation”

(p. 166) would be more likely to have these task-irrelevant responses interfere with their
performance in other testing situations with the same feelings (Sarason & Mandler, 1952).

Work on test anxiety in the 1960’s and 1970’s, along with a great deal of other work in
psychology, took a turn from the behavioralist paradigm toward a more cognitive one. Perhaps
motivated by this shift in the academic culture that incubates such work, subsequent factor analyses
of the Sarason and Mandler’s (1952) Test Anxiety Questionnaire (TAQ) (e.g., Gorsuch,
WRITING ABOUT WORRIES 24

1966; Sassenrath, 1964; Sassenrath, Kight, & Kaiser, 1965) consistently yielded two classes of
factors that work by Sarason had grouped into a single dimension. Work by Liebert and Morris
(1967) was the first to develop distinct lines of inquiry into factors in a cognitive dimension on
the one hand and factors dealing with autonomic arousal (i.e., “emotionality”) on the other.

Liebert and Morris (1967) reasoned that since worry consists of cognitive concerns about
outcomes and the implications of outcomes, the more poorly students expected they would
perform, the more they would experience worry. They also predicted that expressions of
emotionality as characterized by autonomic arousal reflected the degree of perceived uncertainty
in the testing situation. However, they found no support for the latter prediction. Worry and
expectancy were found to be highly correlated while no correlation was found between
emotionality and expectancy. Thus began the trajectory of a literature that examined worry and
emotionality as separate dimensions of test anxiety with differential effects on performance.
Subsequently, Wine’s (1971) Cognitive/Attentional Interference Theory was one of the earliest
that established and defended a bidimensional nature of test anxiety.

Wine (1971) reinforced the support for the bidimensional nature of worry and
emotionality but also extended Liebert & Morris by presenting evidence gathered using Liebert

& Morris’s bidimensional measurement that indicated the cognitive dimension of test anxiety is
more strongly correlated to decrements in performance and offered that test anxiety might be
most aptly defined in terms of attention: decrements in performance are the result of attending to
task-irrelevant thoughts that interfere with task-relevant thought (Wine, 1971).

Despite ample support for the interference model of test anxiety that views test anxiety as
the result of task-irrelevant thoughts that interfere with task-relevant thoughts, the interference
model was challenged throughout the 1980’s by proponents of a skills deficit model that viewed
WRITING ABOUT WORRIES 25

test anxiety as the result of poor performance rather than the cause (Benjamin, McKeachi, Lin &
Hollinger, 1981; Covington, 1985; Kirkland & Hollandsworth, 1980). The locus of attention for
these models, then, becomes the period of test preparation during which encoding and
organization of the information is insufficient, rather than the testing situation itself. However,
this line of research has encountered a number limiting issues. One issue is that students with
strong study-skills and a good track record of performance often report high levels of test anxiety
(Naveh-Benjamin et al., 1987). In addition, interventions for test anxiety that focused on study-
skills or test-taking skills alone found that they had small or non-significant effects (Allen, 1971;
Ergene, 2003; Hembree, 1988; Mitchell, 1972.). These findings undermine the central prediction
of deficit models in that the deficit models hold that students with better skills regarding initial
encoding and organization of material would encounter less anxiety in testing situations. It has
repeatedly been found, however, that students with high levels of study- and test-taking skills
still report high levels of test anxiety and interventions that treat study or test-taking skills alone
do not lead to reductions in test anxiety. The inability of deficit models to explain these findings
has led to the most current research seeded firmly in theoretical frameworks of interference
models.

More recently, Pekrun and colleagues (Pekrun, 2002; Pekrun, Goetz, Titz, & Perry, 2002;
Pekrun, Goetz, Frenzel, Barchfield, & Perry, 2011) have advanced an explanation of test anxiety
that draws on the control-value theory. According to control-value theory, anxiety in testing
situations is a result of the perception of a low amount control over the situation’s outcome and a
high level of subject value in the outcome of the situation (Pekrun, 2006). One of the primary
facets of the control-value theory of achievement emotions is that the theory attempts to move
beyond a focus on test anxiety or even other negative emotions and attempts to account for a
WRITING ABOUT WORRIES 26

fuller treatment of a wider range of emotions including hope, joy, pride, shame, and others
(Pekrun, Goetz, Frenzel, Barchfield, & Perry, 2011; Pekrun, Goetz, Titz, & Perry, 2002). It is
important to point out that Pekrun uses the term emotion in a different way than historical test
anxiety research. Where emotionality has traditionally referred to the autonomic response to
anxiety, Pekrun defines academic emotions as essentially any feelings a student has toward
measured performance and thus casts a broader net into the affective experience of students
others (Pekrun, Goetz, Frenzel, Barchfield, & Perry, 2011; Pekrun, Goetz, Titz, & Perry, 2002).

In summary, Sarason and Mandler’s (1954) theory of test anxiety is considered by many
to be the origin of the modern study of test anxiety. While Sarason and Mandler identified in
their definition of test anxiety aspects of what would come to be known as worry and
emotionality, their initial model viewed test anxiety as a unidimensional construct. Subsequent
work by Liebert and Morris (1967), and later Wine (1971), showed test anxiety to be two
dimensional, with clearly defined cognitive aspects that came to be aggregated under the term

“worry.” These aspects were clearly differentiated from affective and somatic components of
test anxiety and these aspects were aggregated under the term “emotionality.” Research
consistently showed a greater association between worry and performance decrements than
between emotionality and performance decrements. A much greater proportion of research into
test anxiety then examined aspects of worry which were hypothesized to interfere with task-
relevant processing in exam situations. A short-lived but popular interest in identifying a lack of
study and test-taking skills as the source of the worry was dominant for a time in the 1980s, but
this skills-deficit model was unable to account for certain findings. More specifically, the skills-
deficit model was unable to account for consistent findings that students with high study and test
taking skills still experienced test anxiety and that interventions that focused exclusively on
WRITING ABOUT WORRIES 27

developing these skills did not have a significant effect on participants’ test anxiety. With little
evidence to refute interference models of test anxiety, emphasis returned interference models,
especially those provided by Information Processing Theory. Discussed in the next section are
the foundational principles of information processing theory and how these principles continue to
shape research into test anxiety.

The Cognitive Turn

Information Processing Theory

The development of technology and increasing tendency to conceive of the mind and
brain in computational terms can be interpreted as the impetus for the development of
Information Processing Theory, which aims to provide an account of the mental processing of
stimuli, rather than merely the response to it. The initial intent of research grounded in
Information Processing Theory was to provide an account of the internal processes of mentation
not addressed in behaviorist work that centered on the observable external responses to stimuli.
The development of models through the 1950s and 1960s made reference to long-term and short-
term storage of information in their account of memory (cf., Atkinson & Shiffrin, 1968);
however, explanatory models could not account for certain types of experimental data (c.f. Bjork
& Witten, 1974) or offer an explanation that accounted for processing and storage (Dutke &
Stober, 2008). The tripartite model of working memory presented by Baddeley & Hitch (1974)
resolved a number of issues with previous models and with later refinements by Baddeley (e.g.,
1986; 2000) serves as the currently dominant conception of the working memory system.

Baddeley’s (1986, 2001; Baddeley & Hitch, 1974, 1994) model of the working memory
system endorses a tripartite structure. The first component is the phonological loop which serves
in transient storage and manipulation of verbal and acoustic information. The phonological loop
WRITING ABOUT WORRIES 28

consists of two subcomponents: the phonological store, which provides short-term storage of
verbal and acoustic information and an articulatory control system, which serves in the
maintenance of the phonological store. The second component of the working memory system is
the visuospatial sketchpad that serves in transient storage of visual and spatial information.
According to Baddely (2001), the visuospatial sketchpad system is responsible for the storage
and manipulation of visuospatial information. Isolating the dual functions pertaining to visual
information on the one hand and spatial information on the other hand has proven difficult, so
there is less certainty about the potential organization of this system into subcomponents.
Finally, and perhaps most importantly, the third component of the working memory system is
the central executive system. The central executive is considered a modality-free master system
that coordinates the phonological loop system and the visuospatial sketchpad system. The
relevance of each of these systems is discussed at length below.

This model has served as the basis for frameworks that seek to explain the mechanisms
by which test anxiety leads to performance decrements. With continued support for the notion
that factors related to emotionality only effect performance when they are accompanied by worry
(e.g., Deffenbacher, 1980; Morris et al., 1981; Schwarzer 1984), attempts to identify
mechanisms by which test anxiety affect, performance have centered on the functions of these
components and anxiety’s effect on them. Among the most influential of these models that rely
on Baddeley’s conception of the working memory system is Eyesenck’s (1979) Processing
Efficiency theory, later developed into Attentional Control Theory (Eyesenck & Calvo, 1992).

Processing Efficiency Theory and Attentional Control Theory

Processing efficiency theory (Eysenck, 1979; Eysenck & Calvo, 1992) maintains an
important distinction between processing effectiveness and processing efficiency. The former
WRITING ABOUT WORRIES 29

refers to the quality of a task, generally measured in the accuracy of response, while the latter
refers to the relationship between the quality of the performance and the effort expenditure to
achieve that quality (Eysenck, Derakshan, Santos, & Calvo, 2007). Processing efficiency theory
holds that anxiety results in greater decrements in the efficiency of processing than in the
effectiveness of processing. To that extent, processing efficiency theory differs from other
contemporary models of cognitive interference (e.g., Beck, Emery, & Greenberg, 1985) in that it
poses anxiety can serve an important motivational role. That is, individuals prone to high levels
of anxiety in testing situations develop compensatory strategies that might contribute to
decrements in processing efficiency but ultimately promote processing effectiveness (Eysenck
& Calvo, 1992.)

Processing efficiency theory is bound by two primary assumptions. First, it is the worry
component, rather than the emotionality component, of anxiety that is responsible for anxiety’s
effect on performance; Second, the mechanism by which anxiety affects performance is
primarily centered on the central executive system of working memory (Eysenck & Calvo, 1992;
see Derakshan & Eysenck, 2009 for a review of supporting evidence) but also has implications
for the phonological loop as worry typically involves self-dialogue (Borkovec, Alcaine, & Behar,
2004; Borkovec & Inz, 1990; Borkovec & Roemer, 1995; Rapee, 1993).

Methodological and theoretical developments that have allowed for a more fine-grained
analysis of the specific functions of the central executive system of working memory (e.g.,
Miyake et al., 2000; Smith & Jonides, 1999) served as the primary impetus for the revision of
performance efficiency theory into Attentional Control Theory. Based on a more precise
understanding of the function of the central executive system, Attentional Control Theory offers
a more precise explanation of the ways in which anxiety affects working memory. Three
WRITING ABOUT WORRIES 30

functions of the central executive system identified by Miyake et al. (2000) are especially
relevant to attentional control.

The inhibition function of the central executive is characterized by one’s ability to
actively suppress attention to stimuli and is particularly relevant in the inhibition of attention to
task-irrelevant stimuli or responses (Miyake et al., 2000). In a review of approximately 30
studies that examined the effects of anxiety on inhibition during a Stroop test, Derakshan and
Eysenck (2009) found that most of the studies showed high-anxious students are generally
slower than low-anxious students to name the color of the text when the text consists of threat
related words. These findings offer support for the notion that individuals with high levels of
anxiety compared to individuals with lower levels of anxiety are less able to inhibit their
attention toward the threat-related text and identify the color of the text which accounts for the
increase in response time of individuals with high-anxious individuals. It is important to point
out here, however, that the difference in performance between high- and low- levels of anxiety
is in the efficiency of processing and not necessarily the effectiveness of processing.

Additional support for anxiety’s effect on the inhibition function of the central executive
system comes from studies that employ antisaccade tasks. In these studies, participants are
presented with a peripheral stimulus but instructed to look in the opposite direction of the
stimulus. As such, participants must inhibit a reflexive saccade by employing a goal-directed,
top-down, attention process (Posner & Peterson, 1990). Using antisaccade tasks, Derakshan,
Ansari, Hansard, Shoker, and Eysenck (2009) found that anxious individuals took longer to
perform the antisaccade task, but there was no difference in prosaccade latency. In a second
experiment, Derakshan et al. (2009) replicated findings from the first experiment, but found that
the emotional valence of the stimuli also had an effect on antisaccade latency. That is, the
WRITING ABOUT WORRIES 31

latency was greatest when threatening stimuli were used. Collectively, these findings supports
the notion that anxiety is implicated in the inhibitory function of the central executive system
and also provides support to a substantial body of literature that attributes attentional bias toward
threatening material to individuals with high anxiety.

Shifting is defined as the cognitive process of switching between multiple tasks (Miyake
et al., 2000) and is implicated in shifting the locus of attention away from task-relevant stimuli in
the case of distraction or away from task-irrelevant stimuli in the case of returning attention to
task-relevant stimuli. Ansari, Derakshan, and Richards (2008) found significant differences in
performance and latency between high and low anxiety individuals when faced with a task-
switching exercise. Low-anxious individuals actually increased in performance and decreased
latency in task-switching (cf. De Jong, Berendsen, & Cools, 1999) while high-anxious
individuals had no improvement in performance and showed increased latency in task-switching.
Similarly, Derakshan, Smyth, and Eysenck (2009) found in another kind of task-switching
experiment a significant negative effect on performance for high-anxiety individuals but a non-
significant effect on low-anxiety individuals. Taken together these studies provide strong support
for the notion that anxiety has a negative effect on the task switching function of the central
executive system of working memory.

A third function of the central executive system identified by Miyake and colleagues
(2000) is updating, which is more involved in the transient storage of information than in the
control of attention. As such, Attentional Control Theory concerns itself less with the updating
function than with the inhibition and shifting functions. However, recent work in neuroscience
has offered insights into the biological explanation of the mechanisms identified in Attentional
Control Theory.
WRITING ABOUT WORRIES 32

Neural Correlates of Attentional Control

Recent work in cognitive neuroscience (e.g., Bishop, 2007; 2009; Fales, et al., 2008;
Righi et al., 2009; Santos et al., 2011; Savostyanov et al., 2009) has been effective in showing
that anxiety plays a role in enhanced amygdala activation and reduced recruitment of the
dorsolateral prefrontal cortex (DLPFC) and the ventrolateral prefrontal cortex (VLPFC), both of
which are implicated in top-down goal-directed regulation of attention. In addition to this work
that implicates the DLPFC and the VLPFC, recent work (e.g., Hosanagar, et al., 2012; Zhao,
2007) has also been implicated in anxiety responses a network of brain areas referred to as
default mode network (DMN) with activity most consistently elicited during wakeful rest
(Schulman, et al, 1997). Fales et al. (2008) point out that the DMN has been implicated in a
number of anxiety-relevant functions including environmental or visuospatial monitoring,
social and self-referential processing, emotional processing and emotional memory,
autobiographical memory, and mediation of resting-state mental activity (Drevets, 2001;
Greicius, Krasnow, Reiss, & Menon, 2003; Gusnard & Raichle, 2001; Maddock, 1999;
Maddock, Garrett, & Buonocore, 2003; Mazoyer, et al., 2001).

Of particular import with regard to anxiety relevant functions are findings that show
increased activation of DMN accompanied by the onset of anxiety when no cognitive task is required
(Simpson, Drevets, Snyder, Gusnard, & Raichele, 2001; Zald, Mattson, & Pardo, 2002). There is
strong support for the idea that the DMN is functionally anti-correlated with a task-positive network
that controls alerting, orienting, and executive control—one primarily used to engage tasks that
entail attending to external stimuli (Posner & Rothbart, 2005; Smallwood, Fishman, & Schooler,
2007; Stevens, Lauinger, & Neville, 2009.). As activity in the externally concerned task-positive
network increases, activity in the more internally concerned DMN
WRITING ABOUT WORRIES 33

decreases. This could provide for the increase in activity in DMN that accompanies anxiety
possibly preventing the decrease in DMN activity necessary to enable optimal activity in the
task-positive network. It has also been found that individuals with lower levels of trait anxiety
respond to state anxiety by reducing activation in areas affiliated with DMN (Simpson, Drevets,
et al., 2001; Zald et al., 2002). Taken together, these data suggest that levels of state or trait
anxiety could mediate a compensatory strategy of reduced activation in DMN. However, the
increase in DMN activity that accompanies anxiety could potentially inhibit this compensatory
strategy in individuals with high levels of trait anxiety.

In summary, Eyesenck’s (1979) processing efficiency theory holds that high levels of
anxiety inhibit processing efficiency. Further, Eysenck identifies a motivational aspect to test
anxiety in that individuals prone to high levels of anxiety often develop compensatory strategies
such that even with reduced processing efficiency it is still possible to maintain processing
effectiveness. Developments in understanding the function of the central executive (Miyake, et
al., 2000) lead to revising processing efficiency theory into Attentional Control Theory which
focuses on inhibition and shifting. Findings from cognitive psychology about the function of the
central executive are supported by work in neuroscience that shows anxiety plays a role in
reduced recruitment of the DLPFC and VLPFC which are important in top-down goal-directed
regulation of attention.

Insights from neuroscience also show implications for DMN in anxiety-related functions.
Further, the DMN has been found to be functionally anti-correlated with a network that controls
functions that entail attending to external stimuli and that the two essentially toggle on and off.

The literature suggests that anxiety could mediate a compensatory strategy of reduced activation
in DMN regions, but that increases in DMN activity that accompany anxiety could inhibit this
WRITING ABOUT WORRIES 34

strategy in individuals with high levels of anxiety. The following section takes as its primary
purpose the exploration of implications for education of Attentional Control Theory and the
potential role of DMN in the mechanisms described by Attentional Control Theory.

Implications for Interventions

Educational Implications of Attentional Control and Default Mode Network

It is worth noting that in the burgeoning literature surrounding Attentional Control
Theory there does not appear to be any work done that specifically takes the implications of the
theory and applies them to educational practice. Of all of the theoretical predictions that have to
do with the relationship between test anxiety and performance decrements, it does not appear
that there has been any systemic effort to translate the theoretical function of central executive
as identified by Attentional Control Theory to academic testing situations. Similarly, it does not
appear that there has been any attempt to relate empirical findings from studies that examine the
role of DMN in educational interventions. As such, there appears to be a substantial amount of
room for inquiry about how an intervention might be developed given the predictions of
Attentional Control Theory or the implications of work that identifies DMN’s potential role in
cognitive and affective processes engaged in testing situations.

Implications in Health Sciences Education

The author is not aware of any extant literature that presents empirical findings on test
anxiety interventions in the setting of health science education. There is, however, a substantial body
of literature surrounding other aspects of stress, stressors, and anxiety in medical education. A
review of the literature from 1966 to 1999 showed that only 600 studies discussed stress and
stressors in study and practice of health sciences (Shapiro, Shapiro & Schwartz, 2000) while a search
of the same databases for the same time period in other contexts number in the thousands.
WRITING ABOUT WORRIES 35

This is not to say there were no studies regarding stress and anxiety in the health sciences between
2000 and 2011—indeed there were (e.g., Atekin, Karaman, Yigiter Senol, Erdem, Erengin, &
Akaydin, 2001; Cheung & Au, 2011; Kiessling, Schubert, Scheffner & Burger, 2004; Klose Smith,
Peterson, Degenhardt & Johnson, 2007; Markman, Balik, Bruanstein-Bercovitz & Eherenfeld, 2011;
Moffat, McConnacie, Ross, Morrison, 2004; Peterlini, Tiberio, Saadeh, Pereira & Martins, 2002;
Reteguiz, 2006; Sarikaya, Civaner & Kalaca, 2006; Stecker, 2004; Toneatto & Nguyen, 2007;
Vasgegh & Mohammadi, 2007). A pattern of note in these reviews is that the number of studies that
report empirical findings from stress and anxiety interventions is substantially limited. That is, of the
hundreds of articles that discuss stress and anxiety in the health sciences, only 30 studies between
1966 and 2011 report findings from stress and anxiety interventions. For example, of the 12 studies
dealing with stress or anxiety in the health sciences between 2000 and 2011, none include any kind
of intervention. A review of the literature shows only 6 additional studies appear to be have been
published regarding stress and anxiety in the health sciences between 2000 and 2011—a total of 18
studies. The former 12 include no findings from interventions, the latter 6 do—and will be discussed
later in this review.

Even with 30 studies that report empirical findings from stress and anxiety interventions
in the context of the health sciences, it is difficult to discuss findings across the research for a
number of reasons. The remainder of this review will take as its task to discuss these issues.
More specifically, this section of the literature review will take as its focus a discussion of the
multiplicity of kinds of stress and anxiety and the myriad definitions of each; a lack of
standardization of measures of dependent variables; few reports of findings from interventions; a
lack of randomization, utilization of control groups, and follow-up with participants; and a near
WRITING ABOUT WORRIES 36

complete absence of discussion about the biological and psychological mechanisms that would
account for intervention successes and failures.

Even within the specific context of health sciences, there are a number of types of stress
and anxiety that can be examined. While there is some attention to test anxiety in the literature
(e.g., Beggs, Shileds, & Janiszewski Goodin, 2011; Graham, Vettarino, Seifeldin, Singal, 2010;
Ping, Subramaniam & Krishnaswamy; 2008; Powel, 2004; Sanisgiry & Sail, 2005), a great deal
more of the literature deals with more broadly conceived issues of psychological distress and
burnout (Shapiro et al, 2005); mood states (Rozenweig, Riebel, Greeson, Brainard, Hojat &
Mohammadreza, 2003; Finkelstein, Brownstein, Scott & Lan, 2007); and anxiety (Sharif &
Armitage, 2004). Naturally, with only loose relationships between the measured constructs in the
literature that deals with stress and anxiety in the health sciences, it is difficult to reach across
broad swaths of the literature to make comparisons.

Of the 30 studies that reported empirical findings from stress and anxiety interventions in
the health sciences, only 14 studies (Finkelstein et al, 2003; Golden & Rosen, 1975; Graham et
al., 2010; Holtzworth-Munroe, Munroe & Smith, 1985; Kelly, Bradlyn, Dubbert, St. Lawrence,
1982; Kiecolt-Glaser, Glaser, Strain, Stout, Tarr, Holliday & Spelcher., 1986; Nathan, Nixon,
Robinson, Bairnsfather, Allen & Hack, 1987; Palan & Chandwani, 1989; Rozenweig et al.,
2003; Shapiro et al., 1998; Shapiro et al., 2005; Sharif & Armitage, 2004; Whitehouse et al.,
1996;) used control groups and two of these studies were reported in the same paper. Only 10 of
these 14 studies (Graham et al., 2010; Holtzworth-Munroe et al., 1985; Kiecolt-Glaser et al.,
1986; Nathan et al., 1987; Palan & Chandwani, 1989; Shapiro, Astin, Bishop & Cordova, 2005;
Shapiro, Schwartz & Bonner, 1998; Sharif & Armitage, 2004; Whitehouse, et al., 1996)
employed randomization, with results from two of these randomized studies reported in the same
WRITING ABOUT WORRIES 37

paper. Without a control group to compare to a treatment group, however, the utility of the data
is quite limited. Beyond this, when a control group is used but the participation between groups
is not randomized a number of issues with biases can affect the data. A substantial gap in the
literature can be seen with only 10 randomized and controlled studies that report empirical
findings of stress and anxiety interventions in health science education between 1966 and 2011.

A total absence of test anxiety intervention studies and small number of intervention
studies for other kinds of anxiety is further complicated by a lack of standardization of
instruments in the literature. Only two of the six studies conducted in a health science education
setting between 2000 and 2013 that include findings from interventions employ the same
measures: two use the Profile of Mood States (POMS; Rozenweig et al, 2003; Finkelstein et al.,
2007), one uses the Brief Symptom [of Psychological Distress] Inventory (BSI), Maslach
Burnout Inventory (MBI), Perceived Stress Scale, Satisfaction with Life Scale, Self-Compassion
Scale, a ten-point scalar item indicating how meaningful the intervention was, and an open-
ended question used for qualitative analysis (Shapiro, et al., 2005); one used the SCL-90
subscale for Anxiety (Finkelstein et al. 2007); one used the Hamilton Anxiety Scale and
Coopersmith Self-esteem Inventory which were translated into Farsi as well as semi-structured
focus groups (Sharif & Armitage, 2004); one used the Test Anxiety Inventory (TIA) and State-
Trait Anxiety Inventory (STAI; Graham, et al., 2010), and another used the number of subjects
that passed or failed board exams and the number of attempts it took to pass (Powell, 2004). The
lack of standardization of instruments from these studies does not lend itself very easily to
comparison across studies.

None of the studies that report empirical findings from stress and anxiety interventions in
the health sciences provide a framework from which the psychological and biological
WRITING ABOUT WORRIES 38

mechanism of the effects of the intervention can be explained or discussed. This is particularly
interesting given the substantial literature that deals with the relationship between stress and
anxiety and working memory and cognitive load and the literature that deals with task positive,
task negative, and DMN in the brain. Even with the difficulties in comparing the empirical
findings of these interventions in the health sciences, there is a tremendous opportunity to bring
into the discussion of the effects of these interventions the findings of the literature dealing with
working memory, cognitive load, and neural networks. Relatedly, it is worth pointing out that
while a great deal of the literature surrounding working memory, cognitive load, and neural
networks is embedded in both field and laboratory studies, none of the empirical findings about
stress and anxiety interventions in the health sciences stem from laboratory studies in which a
substantial number of variables that could contribute to findings. This is especially important
given the paucity of studies that employ control groups or randomization.

In summary, while there is a relative paucity of discussion about stress and anxiety in the
health sciences compared to other settings, of particular note is the remarkably small number of
studies that report empirical findings of interventions, and of those few studies the number that
employ control groups, and of those even fewer studies, the number of studies that employ
randomization. Beyond that, the lack of a standardization of measures creates an impediment to
comparing findings. Finally, there appears to be an absolute absence of both lab studies and
intervention studies that offer psychological and biological implications of the effects of
interventions and potentially relate the effects to academic achievement. Future research in the
setting of health sciences education should take these aspects of the existing literature into
consideration. Beyond the setting of health sciences education, additional studies lend
themselves to consideration.
WRITING ABOUT WORRIES 39

Recent Interventions Among Other Populations

Hembree (1988) performed a large-scale review of 413 anxiety studies from 1952 to
1986. A large range of categories of interventions was identified, and even wider collection of
subtypes of interventions was reviewed. Consistent with literature reviewed above, test-taking
and study-skills training alone produced small effect sizes while interventions that addressed
worry had a large effect size. Another review of interventions for test anxiety supported these
findings (Ergene, 2003). However, recent work by Ramirez and Beilock (2011) produced rather
remarkable findings through a method not readily discussed in either review.

Drawing on work that showed students who wrote about their feelings about a negative or
stressful experience showed improvements in working memory (Klein & Boals, 2001), Ramirez
and Beilock (2011) had students write about their concerns and worries about an exam
immediately before the exam and found significant improvements in performance. This approach
is similar to those reviewed by Yeager and Walton (2011) who examined interventions targeted
to effect students thoughts, feelings, and beliefs about their educational experiences. Among the
studies reviewed, a number of them included writing interventions intended to develop a
malleable concept of intelligence (Aronson, Fried & Good, 2002), buffer the effects of stereo
type (Cohen, Garcia, Apfel, & Master, 2006; Cohen, Garcia, Purdie-Vaughns, Apfel, &
Brzustoski, 2009; Miyake, Kost-Smoth, Finkelstein, Pollock, Cohen, & Ito, 2010), increase
feelings of belonging (Walton & Cohen, 2007, 2011), align possible selves with racial identity
(Oysterman, Bybee, & Terry, 2006), and increase perceived personal relevance of academic
course content (Hulleman & Harackiewicz, 2009). Each of the studies reviewed resulted in
positive achievement outcomes, although only Ramirez & Beilock (2011), which was published
after the review and thus not included in it, dealt explicitly with test anxiety.
WRITING ABOUT WORRIES 40

It stands to reason that this intervention can be viewed in light of Attentional Control
Theory and work implicating the DMN in cognitive and affective processing. Given that test
anxiety is considered to be a result of the perception of testing situations as threatening and
Attentional Control Theory predicts test anxiety interferes with the central executive functions of
inhibition and switching, it may be the case that the act of articulating thoughts and feelings
about the impending threat allows students to reassess the situation that induces anxious
response. It is consistent with Attentional Control Theory that attending to the kinds of thoughts
that would interfere with task-relevant stimuli and responses on the test before the testing
situation would reduce the amount of potential interference. With a reduction in intrusive
thoughts, there would be a reduction in the demand on the inhibition function. With a reduction
on the inhibition function, it stands to reason there would be more resources to dedicate to task-
relevant thoughts. With less competition between task-relevant and task-irrelevant stimuli, the
incidence of attending to task-irrelevant information would be reduced leading to a decrease in
demand on the shifting function of the central executive system. Taken together, allowing
students the opportunity to explicitly attend to and articulate worry before an anxiety-inducing
testing situation, the anxiety response would be reduced and demand on both the shifting and
inhibition function would be reduced allowing for more resources to be allocated to task-
relevant thoughts.

Beyond Attentional Control Theory, it seems that the intervention employed by
Ramirez and Beilock (2011), and potentially the other interventions reviewed in Yeager and
Walton (2011), is consistent with implications of the work that identifies the role of the DMN
in processing affective information. That is, among students with low levels of test anxiety,
WRITING ABOUT WORRIES 41

increased activation of the outward looking task-positive network are accompanied by
deactivation of DMN in cognitively challenging situations, but in similar situations among
students with high levels of test anxiety the same deactivation of DMN is not seen. Given the
DMN is implicated in interospective processes that have to do with awareness of the self, it may
be the case that a test anxious student’s inability to reduce attention to threats—which in the case
of test anxiety are threats to the ego or self-concept which can be recognized only by turning
attention inward---may be a matter of activating an appropriate network for optimal test
performance. We might predict, then, that the problem of deactivation of the DMN during tests is
attributable to unresolved issues with the cognitive worry induced by the testing situation and
that the very failure to deactivate the inward looking DMN results in a failure to produce optimal
activation of the outward looking task-positive network. As optimal activation of the outward
looking task-positive network is necessary to maximize task-relevant cognitive processes, we can
see how failure to resolve issues of threat and the resultant neural activity that underlay this
failure can inhibit performance in testing situations. Allowing students the opportunity to resolve
some of this threat and thus facilitate deactivation of DMN regions and allow for optimal
activation of the task-positive network which is more conducive to task-relevant cognitive
processing and provide increases in test performance.

However, since Ramirez and Beilock (2011) are the first to employ such an intervention,
any explanatory power of their findings will be increased if these findings are replicated. Given the
very high levels of anxiety in setting of health sciences education, the absence of literature
surrounding test anxiety interventions in the setting of health sciences education, and the potential
for providing explanatory support for Attentional Control Theory and implicating the DMN
network role in processing affective information in educational interventions, the author
WRITING ABOUT WORRIES 42

has decided to attempt to replicate the findings of Ramirez and Beilock (2011) in a health
sciences education setting.

In summary, Attentional Control Theory explains a great deal about how test anxiety
works, but there appears to be little work that applies this theory to interventions for test anxiety
in educational settings. This is coupled with the fact that there does not appear to be any
research that looks at test anxiety in the health sciences education setting. There is a tremendous
body of research that deals with causes, symptoms, and effects of test anxiety and some of that
work does concern itself with the medical education setting; however, the author is not familiar
with any studies that report empirical data from studies that measure outcomes of test anxiety
interventions. However, amongst the relatively few studies that examine test anxiety
interventions in other populations, Ramirez and Beilock (2011) have advanced a research design
that appears, even without explicit mention of Attentional Control Theory or implications of
DMN regions in brain for test anxiety interventions, to support both.

Conclusion

Test anxiety is a specific form of generally subclinical anxiety. While definitions of test
anxiety have remained fairly static since Sarason and Mandler’s (1954) seminal paper, there have
been a number of models that explain relevant mechanisms. Attentional Control Theory relies
heavily on work from cognitive psychology that identifies the component systems of working
memory, specifically within the central executive system of working memory. Insights from
neuroscience support this theory and also implicate recruitment of DMN areas in compensating
for the effects of anxiety. Even with a burgeoning literature that supports this explanation of the
mechanisms involved with test anxiety, there is very little research that moves beyond describing
the causes, symptoms, and effects of test anxiety and offers empirical data that supports the
WRITING ABOUT WORRIES 43

facilitation of test anxiety interventions, especially in the setting of medical education. However,
recent work by Ramirez and Beilock (2011) appears to offer a research design that may provide
explanatory support to both Attentional Control Theory and also to research that implicates the
DMN in compensatory strategies.
WRITING ABOUT WORRIES 44

CHAPTER THREE: METHODOLOGY

Test anxiety is a specific form of subclinical anxiety for which research has long shown
associations with decrements in academic performance. Relative to a large body of research that
examines the causes, symptoms, and effects of test anxiety, there is very little research that
provides empirical findings related to specific interventions, especially in the setting of health
sciences education. The current study aimed to find the effect of expressive writing about
worries on anxiety and performance and to relate findings to existing theory about anxiety and
attentional control as well as any possible role DMN might play in the phenomenon of test
anxiety. This chapter includes the research questions, the hypotheses, and a description of the
research methodology, including the sampling procedure and population, instrumentation, and
procedures for data collection and analysis.

Research Questions

The current research centered on two questions:

1. Is there a difference in exam outcomes for students who write about worries
immediately before an exam and for students who do not?

2. Does writing about worries immediately before an exam reduce self-report survey scores
for state anxiety?

The following are hypotheses regarding the research questions:

1. Writing about worries increases test performance for students.

2. The intervention will reduce state anxiety for high and low test anxiety students, with
a greater reduction for students with greater test anxiety.

Research Design
WRITING ABOUT WORRIES 45

The methodology for this study is adopted from Ramirez and Beilock (2011), which the
current study seeks to extend by including measurement of state anxiety and by administering the
intervention to two populations not included in the original study. As the setting and context was
somewhat different, there were necessary modifications to the research design.

At Time 1, the study was introduced in class. For Sample 1, at this time, demographic
information questionnaires, cognitive test anxiety scales, and trait anxiety inventories were
distributed to students who were asked to fill them out and submit them at the end of the
introduction to the study. For Sample 2, a link to a Qualtrics survey that included the same
instruments was provided. At Time 2 for both samples, immediately before the second test of the
semester, experiment packets were randomly distributed to participants. Students received one of
two packets: either a treatment group packet or a control group packet. Included in the packet for
the treatment group was the state anxiety scale, which was completed immediately before the
intervention, a writing prompt that asked students to write for 10 minutes about their worries and
concerns regarding the test, and an additional copy of the state anxiety scale that was completed
immediately after the intervention. At the same time, the control group was provided with a
packet that contains the state anxiety scale to be completed immediately before the intervention,
a prompt that asked students to think for 10 minutes about what they believe will not be covered
on the test, and an additional copy of the state anxiety scale that was completed immediately
after the intervention. Immediately following administration of the postintervention state anxiety
scale, students commenced with the exam. In Sample 1, after the exam was graded, university
identification numbers were coded by a double-blind referee. In Sample 2, university
identification numbers were used to identify participants. In both studies aggregate data were
analyzed.
WRITING ABOUT WORRIES 46

Population and Sample

The populations of interest are undergraduate students in a dental hygiene program
(Sample 1) and undergraduate students in a non-dental hygiene program (Sample 2). In Sample
2, participants were recruited from an undergraduate learning skills and motivation course.
While there are some differences between these samples and students at the K-12 level, as well
as differences between health science disciplines and pre- or postbaccalaureate programs, the
biological and psychological mechanisms this intervention targets should open the possibility of
leaving findings generalizable to a much wider population of individuals dealing with worry
that inhibits performance, preparation for performance, or constructive reflection on
performance. Even though the setting is health sciences education and test anxiety is largely
associated with academic settings, the findings should be generalizable beyond health sciences
education and even into nonacademic settings where social or performance anxiety is relevant.

Undergraduates entering a dental hygiene program as juniors (N = 31) and continuing
dental hygiene students in their senior year (N = 26) made up Sample 1. Participants in Sample 1
were 89% female; 35% were White, 2% were African-American, 19% were Hispanic/Latino,
12% were Asian , 5% were Native Hawaiian or Pacific Islander, 27% were Other or two or more
races. Students, enrolled in a mandatory learning skills and motivation course (N = 101) at
University of Southern California made up Sample 2. Participants in Sample 2 were 62%
female; 59% were White, 11% were African-American, 13% were Hispanic/Latino, 8% were
Asian , 1% were Native Hawaiian or Pacific Islander, 8% were Other or Two or more races.
Table 1 presents demographic information for Sample 1 and Sample 2.
WRITING ABOUT WORRIES 47

Table 1

Demographic Information for Sample 1 and Sample 2

Sample 1 Sample 2

n 57 101
Male 6 62
Female 51 39
White 20 59
African-American 1 11
Hispanic/Latino 11 13
Asian 7 8
Hawaiian/Pacific Islander 3 1
Other/2 or More Races 15 9

Instrumentation

Instruments employed in the study included a demographics questionnaire, the
Cognitive Test Anxiety Scale, the Costelleo-Comrey Anxiety Inventory, the short form of
the State-Trait Anxiety Inventory state anxiety subscale, an intervention writing prompt, and
a control group writing prompt.

Demographics questionnaire. The demographics questionnaire was developed for this
study and includes basic demographic information including age, gender, ethnicity, and student
ID number (Appendix 1).

Cognitive test anxiety inventory. Propensity for cognitive test anxiety, or worry, was
measured with the Cognitive Test Anxiety Inventory, which is available in the public domain and
was found to have a high level of internal consistency, α = .91 (Cassady & Johnson, 2002).
WRITING ABOUT WORRIES 48

In the current study, similarly high levels of internal consistency were found, α = .89. Students
responded to 27 statements such as “I tend to freeze up on things like intelligence tests and final
exams” and “My mind goes blank when I am pressured for an answer on a test” on a four point
scale where A = Not at all typical of me, B = Only somewhat typical of me, C = Quite typical of
me, and D = Very typical of me (Appendix 2).

Trait anxiety scale. Trait anxiety was measured by the Costello-Comrey Anxiety Inventory
(Costello & Comrey, 1967), which is available in the public domain and was found to have a high
level of internal consistency, α = .90 (Serovich, Kimberly, Mossack & Lewis, 2001).

In the current study, similarly high levels of internal consistency were found, α = .87. Students
respond to nine statements such as “I get rattled easily” and “When faced with excitement or
unexpected situations, I become nervous and jumpy” on a 9-point scale ranging from
Always/Absolutely Agree to Never/Absolutely Disagree (Appendix 3).

State anxiety scale. State anxiety was measured by Marteau and Bekker’s (1992) short
form of the State-Trait Anxiety Inventory state-anxiety subscale, which is available in the
public domain and was found to have a high level of internal consistency, α =.82. In the current
study, similarly high levels of internal consistency were found, α = .85 for preintervention
administration and, α = .87 for postintervention administration. Students responded to six
statements such as “I feel tense” and “I am relaxed” on a 4-point scale where 1 = Strongly
Disagree, 2 = Somewhat Disagree, 3 = Somewhat Agree, and 4 = Strongly Agree (Appendix 4).

Intervention group and treatment group writing prompts. Finally, prompts for
treatment and control group have been adopted directly from Ramirez and Beilock (2011). The
intervention group received a prompt that asks them to write as expressively as they can for 10
minutes about their concerns and worries about the exam (Appendix 5). The control group
WRITING ABOUT WORRIES 49

received a prompt that asks them to think as factually as they could for ten minutes about
what they did not believe would be on the test (Appendix 6).

Data Collection

Sample 1

Data collection began at Time 1 where students received a brief introduction to the study
at which demographic questionnaires, cognitive test anxiety scales, and trait anxiety scales were
distributed to students. Students were instructed to complete each and submit them at the end of
the introduction to the study

Time 2 occurred during the final exam for 1
st
-year dental hygiene students and during
the second of eight exams for 2
nd
-year dental hygiene students. There was no intervention during
the 1
st
-year dental hygiene mid-term or the first test of the course for 2
nd
-year dental hygiene
students. At Time 2, students arrived at the location of their exam and were asked to seat
themselves. A brief set of instructions was then read and control and treatment packets were
randomly distributed. The packets included the following items: a cover sheet with a printed
version of the instructions read by the proctor, a pre-intervention copy of the state anxiety
measurement, a copy of either the intervention writing prompt or the control group writing
prompt with writing paper and writing instrument included, and a post-intervention copy of the
state anxiety measurement.

Students were instructed to remove the packet from the envelope, complete the pre-
intervention state anxiety inventory, read the writing prompt they were provided, perform the
writing, complete the post-intervention state anxiety scale, and return the packet to the envelope.
Instructions were specific that in the event a student finishes before the allotted time expires,
they were to sit quietly and wait for the proctor to announce collection of the envelopes. Once
WRITING ABOUT WORRIES 50

all envelopes were collected, students began their exam. After exams were graded, the course
coordinator provided grades to the experimenter.

Sample 2

Sample 2 varied only slightly in data collection. For Sample 2, data collection began at
Time 1 when the study was introduced to the class subsequent to the midterm and prior to the
final. Rather than distributing hard copies of the demographics questionnaire, trait anxiety
survey, and cognitive test anxiety survey, all students enrolled in the course to which the
introduction to the study was given were emailed a link to a Qualtrics survey that included all
three instruments. Time 2 occurred just prior to the final exam for the course and data collection
for Sample 1 and Sample 2 were identical for Time 2. Table 2 presents the timeline of data
collection between samples.

Table 2

Timeline of data collection between samples

Sep Oct Nov Dec
Sample 1
1
st
-year Time 1 Midterm Final/
Students Time 2
2
nd
-year Exam1 Time 1 Exam2/ Exam3 Exam4 Exam5 Exam6 Exam7 Exam8
Students Time 2
Sample 2 Midterm Time 1 Final/
Time 2

Quantitative Analysis

To answer the first research question regarding the interaction between intervention and
exam performance, ANCOVA analysis was used to examine the effect of condition differences,
as independent variable, on exam performance, as dependent variable, controlling for prior exam
performance, as covariate. Subsequently, regression analysis was used to examine if total word
WRITING ABOUT WORRIES 51

count of the writing produced during the intervention, as independent variable, predicted
performance, as dependent variable.

To answer the second research question regarding the interaction between the
intervention and state anxiety levels, T-tests were computed to determine if mean scores of
preintervention state anxiety differed from postintervention state anxiety. Subsequently,
regression analysis was used to see if the level of preintervention state anxiety, as independent
variable, significantly predicted the amount of change, as dependent variable, between
preintervention state anxiety and postintervention state anxiety. Finally, regression analysis was
used to see if the level of cognitive test anxiety, as independent variable, significantly predicted
the amount of change, as dependent variable, between preintervention state anxiety and
postintervention state anxiety.

Analysis of Writing Samples

The Linguistic Inquiry and Word Count (Pennebaker, Booth, & Francis, 2007) was
used to analyze the content of the writing performed during the intervention. Proportion of text
that fell into several word categories (Anxiety, Affect, Positive Emotion, Negative Emotion,
Cause, Insight, & Inhibition) was calculated and then correlations were conducted between
mean proportion text in each word category, exam performance, and change between
preintervention and postintervention state anxiety
WRITING ABOUT WORRIES 52

CHAPTER FOUR: RESULTS

This chapter presents the statistical outcomes for the previously presented research

questions:

1. Is there a difference in exam outcomes for students who write about worries immediately
before an exam and for students who do not? It is hypothesized that writing about worries
immediately before an exam will be associated with high exam performance.

2. Does writing about worries immediately before an exam reduce self-report survey scores
for state anxiety? It is hypothesized that writing about worries immediately before an
exam will reduce self-report survey scores for state anxiety.

Intercorrelations

Correlational analyses of variables in Sample 1 yielded positive associations between,
trait anxiety and intervention word count, r = .45, p = .024, change between pre- and
postintervention state anxiety, r = .331, p = .027, cognitive test anxiety, r = .517, p < .001, and
preintervention state anxiety, r = .585, p <.001. Positive associations were also found between
preintervention state anxiety and change between pre- and postintervention state anxiety, r =

.609, p <.001, cognitive test anxiety, r = .388, p = .006, and trait anxiety, r = .585, p < .001.
Table 3 presents intercorrelations between variables in Sample 1.

Correlational analyses of variables in Sample 2 yielded positive associations between trait
anxiety and cognitive test anxiety, r = .509, p < .001, and preintervention state anxiety, r =

.452, p <.001. Table 4 presents intercorrelations between variables in Sample 2.
Correlational analyses across both samples yielded positive a positive association

between total word count from the writing intervention and exam performance, r = .282, p =

.030, and a marginally significant positive association between word count and cognitive test
WRITING ABOUT WORRIES 53
Table 3
Intercorrelations between variables for Sample 1

Perf. Count Pre-Post CTA Trait State

Perf. -- -- -- -- -- --
Count .189 -- -- -- -- --
Pre-Post .046 .211 -- -- -- --
CTA -.053 .255 .257 -- -- --
Trait -.062 .450
*
.331
*
.517
**
-- --
State -.110 .220 .609
**
.388
**
.585
**
--

Notes. Perf. = Exam Performance. Count = Word count. Pre-Post = Change from Pre- to
Postintervention State Anxiety. CTA = Cognitive Test Anxiety. Trait = Trait Anxiety. State =
Preintervention State Anxiety.

*p ≤ 0.05 level; **p ≤ .01 level.

anxiety, r = -.265, p = .058 Additional positive associations were found between change between
pre- and postintervention state anxiety and cognitive test anxiety, r = .370, p < .001, trait anxiety,
r = .234, p = .015, and preintervention state anxiety, r = .186, p = .042. Finally, a positive
association was found between cognitive test anxiety and trait anxiety, r = .511, p < .001. Table
5 presents intercorrelations between variables across both samples.

Research Question 1: Exam outcomes

Using ANCOVA analysis, it was found that that in Sample 1, F(1, 45) = .363, p = .550,
Sample 2, F(1, 69) = .151, p = .699, and across both samples, F(1, 117) = .002, p = .964, there
was no significant interaction between condition differences and exam performance using prior
WRITING ABOUT WORRIES 54
Table 4
Intercorrelations between variables for Sample 2

Perf. Count Pre-Post CTA Trait State

Perf. -- -- -- -- -- --
Count .164 -- -- -- -- --
Pre-Post -.062 -.193 -- -- -- --
CTA .099 -.228 .110 -- -- --
Trait .142 -.157 .005 .509
**
-- --
State -.028 -.166 .172 -.239 .452
**
--

Notes. Perf. = Exam Performance. Count = Word count. Pre-Post = Change from Pre- to
Postintervention State Anxiety. CTA = Cognitive Test Anxiety. Trait = Trait Anxiety. State =
Preintervention State Anxiety.

*p ≤ 0.05 level; **p ≤ .01 level.

exam score as a covariate. However, regression analysis showed that across both samples, for
students who participated in the treatment condition, total word count of writing produced
during the intervention significantly predicted exam score, β = .288, t(56) = 2.231, p = .030.

Research Question 2: State anxiety

T-tests were computed to determine if mean scores of preintervention state
anxiety differed from postintervention state anxiety for participants in the treatment condition in
Sample 1, Sample 2, and across both samples. In Sample 1, postintervention state anxiety was
significantly lower than preintervention anxiety, t(21) = 2.193, p = .040, but there were no
significant differences between preintervention state anxiety and postintervention state anxiety
WRITING ABOUT WORRIES 55
Table 5
Intercorrelations between variables across Sample 1 and Sample 2

Perf. Count Pre-Post CTA Trait State

Perf. -- -- -- -- -- --
Count .282
*
-- -- -- -- --
Pre-Post -.155 -.187 -- -- -- --
CTA -.119 -.265 .370
**
-- -- --
Trait .003 -.034 .234
*
.511
**
-- --
State .041 .105 .186
*
-.058 -.044 --

Notes. Perf. = Exam Performance. Count = Word count. Pre-Post = Change from Pre- to
Postintervention State Anxiety. CTA = Cognitive Test Anxiety. Trait = Trait Anxiety. State =
Preintervention State Anxiety.

*p ≤ 0.05 level; **p ≤ .01 level.

amongst participants in the treatment conditions for Sample 2, t(37) = -1.782, p = .083, or across
both studies, t(58) = -.049, p = .961.

In addition, regression analysis showed that in Sample 1 and across both , for students
who participated in the treatment condition, the level of preintervention state anxiety
significantly predicted the amount of change between preintervention state anxiety and
postintervention state anxiety, β = .593, t(20) = 3.210, p = .005, and β = -.432, t(57) = -3.580, p

= .001, respectively. However, in Sample 2, for students who participated in the treatment
condition, the level of preintervention state anxiety did not significantly predict the amount of
WRITING ABOUT WORRIES 56

change between preintervention state anxiety and postintervention state anxiety, β = .194, t(36)
= 1.171, p = .249.

Finally, regression analysis showed that across both samples the level of cognitive test
anxiety significantly predicted the amount of change between preintervention state anxiety
and postintervention state anxiety, β = .370, t(105) = 4.059, p < .001.

Analysis of Writing Samples

The Linguistic Inquiry and Word Count (Pennebaker, Booth, & Francis, 2007) was
used to analyze the content of the writing performed during the intervention. Proportion of text
that fell into several word categories (Anxiety, Affect, Positive Emotion, Negative Emotion,
Cause, Insight, & Inhibition) was calculated and then correlations were conducted between
mean proportion of word categories, exam performance, and change between preintervention
and postintervention state anxiety. No statistically significant correlations (all ps > .05) were
found between any of the categories and exam performance or change between preintervention
and postintervention state anxiety.
WRITING ABOUT WORRIES 57

CHAPTER FIVE: DISCUSSION

This chapter will present a discussion of the findings presented in the last chapter. The
purpose of the current research was to extend text anxiety literature beyond identifying causes,
symptoms, and effects of test anxiety by providing empirical data about effective interventions
for test anxiety. More specifically, the purpose of the current study was to see if writing about
worries regarding the outcome of an exam immediately before the exam reduces levels of state
anxiety and increases performance. Additionally, group differences in levels of anxiety were
examined between first- and second-year dental hygiene students and between dental hygiene
students and other students to see if differences exist in levels of anxiety. As such, the research
questions that guided the current study were as follows:

1. Is there a difference in exam outcomes for students who write about worries
immediately before an exam and for students who do not?

2. Does writing about worries immediately before an exam reduce self-report survey scores
for state anxiety?

The following are hypotheses regarding the research questions:

1. Writing about worries increases test performance for students.

2. The intervention will reduce state anxiety for high and low test anxiety students, with
a greater reduction for students with greater test anxiety.

Hypotheses About Performance and State Anxiety

It was hypothesized that writing before an exam would increase performance. As
discussed previously, this intervention can be viewed in light of Attentional Control Theory and
work implicating the DMN in cognitive and affective processing. Given that test anxiety is
WRITING ABOUT WORRIES 58

considered to be a result of the perception of testing situations as threatening and that Attentional
Control Theory predicts test anxiety interferes with the central executive functions of inhibition
and switching, it may be the case that the act of articulating thoughts and feelings about the
impending threat allows students to reassess the situation that induces anxious response. Further,
it is consistent with Attentional Control Theory that attending to kinds of thoughts that would
interfere with task-relevant stimuli and responses on the test before the testing situation would
reduce the amount of potential interference during the testing situation. With a reduction in
intrusive thoughts, there would be a reduction in the demand on the inhibition function. With a
reduction on the inhibition function, it stands to reason there would be more resources to
dedicate to task-relevant thoughts. In addition, with less competition between task-relevant and
task-irrelevant stimuli, the incidence of attending to task-irrelevant information would be
reduced leading to a decrease in demand on the shifting function of the central executive system;
therefore, it was hypothesized that allowing students the opportunity to explicitly attend to and
articulate worry before an anxiety-inducing testing situation, the anxiety response would be
reduced and demand on both the shifting and inhibition function would be reduced allowing for
more resources to be allocated to task-relevant thoughts. Reallocation of those resources, then,
should lead to an increase in performance on the test.

Beyond Attentional Control Theory, it would seem that the intervention employed by
Ramirez and Beilock (2011), and potentially the other interventions reviewed in Yeager and
Walton (2011),is consistent with implications of the work that identifies the role of the DNM
in processing affective in formation. That is, among students with low levels of test anxiety,
increases activation of the outward looking task-positive network are accompanied by
deactivation of DMN in cognitively challenging situations, but in similar situations among
WRITING ABOUT WORRIES 59

students with high levels of test anxiety the same deactivation of DMN is not seen. Given the
DMN network is implicated in introspective processes that have to do with awareness of the self,
it may be the case that a test anxious student’s inability to reduce attention to threats—which in
the case of test anxiety are threats to the ego or self-concept which can be recognized only by
turning attention inward—may be a matter of activating an appropriate network for optimal test
performance. We might predict that the problem of deactivation of the DMN during tests is
attributable to unresolved issues with the cognitive worry induced by the testing situation and
that very failure to deactivate the inward looking DMN results in a failure to produce optimal
activation of the outward looking task-positive network. As optimal activation of the outward
looking task-positive network is necessary to maximize task-relevant cognitive processes, we can
see how failure to resolve issues of threat and the resultant neural activity that underlay this
failure can inhibit performance in testing situations. Allowing students the opportunity to resolve
some of this threat and thus facilitate deactivation of DMN regions and allow for optimal
activation of the task-positive network which is more conducive to task-relevant cognitive
processing and provide increases in test performance.

Contrary to predictions, data from the current study showed a significant reduction in
state anxiety in only one sample and did not show a significant increase in performance. This
raises the question of why the data do not offer greater support for the hypotheses about state
anxiety and performance. Several possible explanations emerged as contextual factors were
analyzed.

Processing Efficiency Versus Processing Effectiveness

One of the primary assumptions of Processing Efficiency Theory, and of Attentional
Control Theory which Processing Efficiency Theory gave rise to, is that anxiety results in greater
WRITING ABOUT WORRIES 60

decrements to the efficiency of processing than to the effectiveness of processing. That is,
decrements in the relationship between the quality of the performance and the effort expenditure
to achieve that quality are more pronounced than decrements in the quality of the processing,
generally measured in the accuracy of responses (Eysenck, Derakshan, Santos, & Calvo, 2007).
While the intervention may have had little effect on the ultimate quality of processing, indicated
by exam score, there still may be a significant relationship between the intervention and the
efficiency of processing. The design of the current study did not allow for fine-grained analysis
of this question; however, the relationship between word count and exam score seems to offer
some insight into this relationship.

Word Count and Content of Writing as Indicator of Fidelity to Treatment

For participants who wrote about worries, total word count ranged from 38-268—a
difference of more than 700%. Given that the theoretical framework that informs this study
holds that the opportunity to write about worries before an exam affords students the
opportunity to reassess the situation that induces anxious response, students that invest only a
minimal level of fidelity to the treatment should not benefit in the same way as those that invest
a greater level of fidelity to the treatment. As Yeager & Walton (2011) point out, interventions
targeting student’s students thoughts feelings and beliefs are not magic. Rather, these types of
interventions intend to elicit specific mechanisms and derivation from protocol in ways that lead
to a failure of mechanism in question to be elicited not lead to outcomes that demonstrate
predicted benefits. It appears to be possible that there may have been an insufficient number of
participants in the treatment group that produced a sufficient amount of writing to affect the
efficiency of processing enough to affect its effectiveness.

For example, the entirety of one participant’s writing was as follows:
WRITING ABOUT WORRIES 61

I am nervous to take the exam and taking this is making me more nervous and antsy. I
usually don’t like to focus on other things right before an exam. Having 3 midterms this
week is very overwhelming.

The word count of the above sample is in sharp contrast to the following sample:

Not gonna lie--I'm pretty nervous about this exam. I have a tendency to get this way
before any assignment. I am a very stressed person and I am REALLY SCARED to take
this exam. Did I study enough? Did I review to correct material? Some people in the
class just mentioned something and I literally have NO idea what they are talking about!
Oh geez. I'm not sure if this is supposed to be train of thought, but that's what I'm making
it. I would really like to do well on this exam. I have this weird complex where I feel like
I need to prove to my professors that I'm a dedicated student, by getting really good
grades. This leads me to get stressed over the most absurd thins. For example: I had a
homework assignment a few weeks back where I had to write a poem in French. I am in
French 1...so, the extent of my knowledge in French is fairly limited. But, I love poetry
and have always taken an interest in it, so I'm sitting there in Leavey, writing this French
poem, when it starts to hit me that the complexity and beauty (two major factors of
poetry) of my poem are going to be really bad. Like, REALLY bad. Then my heart
started racing because I thought my French teacher would judge me for the simplicity of
my poem--she definitely doesn't care if I'm not Walt Whitman. This led to my panic
attack in Leavey over a poem -- a poem?! So, yes, I'm very stressed about this exam.

While both samples indicate a degree of anxiety about the exam, it may well be the case that the
author of the first sample was so nervous that it impeded his or her ability to produce a greater word
count inhibiting the effect of the intervention. Further, in the second sample, while some of
WRITING ABOUT WORRIES 62

the content does deal with anxiety about the exam, there is also a certain amount of discussion
about worries outside of the exam. A reduction of anxiety outside of the context of the exam
may have been present, but it may be the case that the intervention did not serve to reduce
anxiety regarding the exam. This interaction could account for the author of the second sample
reporting a higher level of state anxiety after the intervention than before. An alternative
explanation is discussed below.

In sum, the intervention may well have affected processing efficiency in significant
ways, but lack of fidelity to the treatment within the control group may have prevented benefits
of increasing processing efficiency to manifest as improved performance as measured by exam
score. Additional support for this idea comes from results of analyses with regard to the second
research questions which asked if writing about worries reduces state anxiety.

Preintervention Level of State Anxiety

In Sample 1 and across both samples, it was found that the level of preintervention state
anxiety significantly predicted the postintervention reduction in state anxiety. That is, the greater
the level of anxiety when beginning writing about worries, the lower the subsequent level of
anxiety. This is consistent with previous research including Ramirez and Beilock (2011), on
which the design of the current research is based. In general, students with a higher level of state
anxiety stand to encounter more worry-based task-irrelevant thought. As the intervention was
intended to allow for reappraisal of the anxiety-inducing situation, and thus to allow for
attenuation of worry-based task-irrelevant thought, the greater the level of worry-based task
irrelevant thought, the greater the potential to reduce the level of worry-based task-irrelevant
thought. Consistent with control-value theory, if a student felt like he or she had a substantial
amount of control over the outcome of the exam, either based on pervious performance or on
WRITING ABOUT WORRIES 63

preparation, or if the student did not hold a particularly high level of importance for the outcome
of the exam, there would be only a minimal anxiety response leaving the intervention with only
the possibility of very small changes in anxiety, if any (Peckrun, 2006). It may be the case then
that beginning the intervention with a particularly low level of state anxiety provides for a sort of
reverse ceiling effect such that low levels of state anxiety cannot be further lowered, while higher
levels of state anxiety lend themselves to the kind of reduction that might be expected as per the
reasoning behind the hypotheses. This was not the case, however, in Sample 2, which had the
highest mean preintervention state anxiety score. What might account for this anomaly given that
we would predict that greater the preintervention state score, which was highest in Sample 2, the
greater the difference between pre- and postintervention state anxiety?

One possible explanation may be drawn from Pani, Askar, Al Mohrij, and Ohali (2011)
which showed a negative correlation between perceived academic stress and salivary cortisol
levels. Because salivary cortisol is a marker of stress, it could be predicted that when perceived
academic stress was the high, so too should be levels of salivary cortisol, but this was not case:
for some students, when a biological marker of stress reported low levels of stress, they reported
high anxiety; likewise, when a biological marker of stress reported high levels of stress, they
reported low levels of anxiety. Findings from the current study and Pani et al. (2011) could
potentially be explained by metacognitive interpretations of affect. Interpreting arousal in
positive ways such that it was viewed as assistive in navigating an anxiety-inducing situation as
opposed to than as a sign of impending doom, was associated with increased cardiac efficiency,
lowered vascular resistance and decreased attentional bias in one study (Jamieson, Nock, &
Mendes, 2011) and increased performance for students in a stereotype induction in another
study (Schmader, Forbes, Zhang, & Mendes, 2009). It could be the case that students who
WRITING ABOUT WORRIES 64

initially experienced a high degree of anxiety regarding the exam interpreted that anxiety as
something that would help them on the exam, which led to a reduction in anxiety. In the same
way, students who initially experienced a low degree of anxiety regarding the exam could have
interpreted that anxiety as sign that they entering into a threatening situation which could have
led to an ultimate amplification of the anxiety. Additional explanation of unexpected findings
regarding levels of state anxiety may be found in consideration of levels of cognitive test anxiety,
which should predict levels of state anxiety prior to the exam.

Level of Cognitive Test Anxiety

While participants in Sample 2 had the highest level of state anxiety, they also had the
lowest levels of cognitive test anxiety. This could be explained in a number of ways. One
assumption about administering the state anxiety survey immediately before an exam is that
reports of levels of state anxiety would reflect anxiety regarding the exam. However, given that
participants in Sample 2 reported a lower level of cognitive test anxiety than did participants in
Sample 1, and given that the intervention did not reduce state anxiety for these participants in a
statistically significant way, it could be the case that the high level of state anxiety experienced
by students prior to the intervention was not necessarily associated with taking the exam.
Indeed, cognitive test anxiety significantly predicted the difference in pre- and postintervention
state anxiety across both samples. If it was the case that the state anxiety experienced by
participants in Sample 2 was not a result of the impending exam, no matter what level of
fidelity to the intervention students demonstrated, the intervention would not be likely to reduce
whatever non-test-related anxiety students were experiencing. Preliminary analyses of the
content of the writing support the notion that these students were not especially anxious about
the exam.
WRITING ABOUT WORRIES 65

This interpretation is consistent with control-value theory in so far as the protocol of the
intervention assumes that given the proximity of the exam, and anxiety a student feels would come
from the exam itself. However, if a student coming into the intervention was weighed down by a
sense of low controllability with regard to another situation in which they held a high subjective
value for the outcome, we would expect a similar anxiety response (Pekrun, 2002). Since the seat of
the anxiety would be in a situation outside of the test, writing about worries about the exam may not
allow for the same kind of reappraisal that would lead to a reduction of the anxiety. In such a
situation, the intervention could still be functioning effectively, but the context of the administration
of the intervention would be such that the intervention would unlikely be able to produce any
measurable change in anxiety or performance.

Explicit Instruction on Compensatory Strategies

Perhaps also accounting for differences in both performance and anxiety between
samples is the fact that emotional and cognitive self-regulation and dealing with stressors was
part of the course from which Sample 2 was recruited. While none of the approaches to stress or
anxiety in the course dealt with expressive writing, it may very well be the case that the
proximity of exam review—which would have entailed review of the course content that dealt
with strategies for handling stressful situations in academic contexts—to the intervention
allowed for intervention-irrelevant amelioration of anxiety. In the same vein, it may be worth
noting that a substantial portion of Sample 2 was made up of student-athletes, many of which
compete on teams that contend at the level of national championships. Indeed, over the course of
the semester, at least some of the students in the class were on an undefeated NCAA national
championship winning team. Many of these students may have had previous exposure to
WRITING ABOUT WORRIES 66

strategies for dealing with performance-based anxiety over the course of their athletic careers,
which we would expect to exert a significant influence on the outcome of the intervention.

Limitations of the Study

There are several limiting factors to study. The first is that the design of the study
does not take into account all confounding variables that could contribute to performance. The
study makes no attempt to measure or account for study or test-taking skills which have been
shown to effect performance. There are also myriad other factors that will not be measured or
controlled for including previous experience with the content of the test including having
participants who at the time of the test are repeating the curriculum, as well as a number of
personal situations or characteristics that contribute to performance. A second limitation is the
study’s reliance on self-report survey data. The self-report survey measures cannot, and are not
intended, to provide an error-proof, objective measurement of the constructs they measure. The
surveys are assumed and intended to be valid and reliable scales of the construct that enable
comparative discussion. Finally, different testing situations may induce varying levels of anxiety
that effect performance to varying degrees. For any number of reasons, a student may have
experienced a low level of test anxiety during the administration of Exam 1 and a high level of
test anxiety in Exam 2. As such, care must be taken in analyzing the findings and drawing
conclusions.

Sample 1 was made up of first- and second-year students in a dental hygiene program.
The exam before which the intervention took place was not the same for the two classes of
students. Differences in workload, maturity in the program, and the exam itself are likely to exert
an effect on the outcome of the intervention. Additionally, Sample 2 was made up of students
enrolled in a mandatory learning skills and motivation course. As such, we might
WRITING ABOUT WORRIES 67

expect substantial differences in motivational orientation toward the course and thus the
exam that served as impetus for the intervention.

Further, the sample that made up Sample 2 took the exam prior to a very different kind
of exam: rather than a forced-answer multiple choice test, participants in Sample 2 took an exam
consisting of primarily open-ended short response questions. Exams were graded a cohort of
eight different graduate teaching assistants. Inter-rater reliability was not assessed. Asymmetry
in sample and exam type and an absence of inter-rater reliability could potentially have
implications for the validity of the statistical analyses on which the finding of the current study
are based. This issue is discussed further with regard to implications for future research.

Finally, there were a large portion of students in Sample 1 for whom English is not the
primary language. As a post hoc consideration, specific data regarding primary languages were
not collected for the study; however, program administration suggests that a significant portion
of the sample is nonnative English speakers. Anxiety about writing, as opposed to anxiety about
the exam, could have been a confounding variable.

Delimitations

The study delimits the issue of confounding variables by measuring both trait anxiety
and propensity toward cognitive test anxiety weeks before the intervention. In addition, state
anxiety was measured immediately before and immediately after the intervention. Further,
analysis of some data indicative of confounding variables (e.g., performance on previous exam)
was controlled for in statistical analysis of the results of the intervention.

The issue surrounding the use of self-report survey data was delimited by employing only
standardized self-report survey measures available in the public domain for which strong validity
and reliability have been demonstrated. As the literature is populated with studies that use
WRITING ABOUT WORRIES 68

unstandardized self-report survey measures, often constructed by the authors of the studies
that use them, using instruments available in the public domain lends itself to amelioration of
an overarching issue in the literature with standardization of instruments and also increases the
validity of the findings as the instruments have been demonstrated to show strong validity and
reliability.

Delimitation of the issue of differential loads of anxiety in different evaluation situations
to which the evaluative situation that includes the intervention will be compared is addressed in
part by gathering data about trait anxiety and cognitive test anxiety, both of which are
considered to be fairly stable relative to state anxiety. In addition, consideration to confounding
variables (e.g., prior performance) was given in the statistical analysis of data pertaining to the
performance on which confounding variables might exert influence.

Implications for Students, Instructors, Support Personnel, and Administrators

The current study situates itself at the intersection of several lines of research. Among the
literature that examines test anxiety, primary foci are the causes, symptoms, and effects of test
anxiety. Relative to the number of studies that examine causes, symptoms, and effects of test
anxiety, there are relatively few studies that examine empirical evidence from field studies of
interventions. One of the primary aims of the current study is to contribute suggestions for
practice to the discussion regarding what can be done to assist students who experience
deleterious effects of test anxiety.

While the findings warrant further investigation, they also offer modest support for the
benefits of writing about worries before an exam as an intervention for test anxiety. The
simplicity of the nearly cost-free intervention lends itself to easy adoption within restrictive
classroom contexts and findings across both Ramirez and Beilock (2011) and the current study
WRITING ABOUT WORRIES 69

support an association between the intervention and performance outcomes as well as anxiety
levels. It is advisable to invite students to write openly, freely, and expressively about their
worries about an exam in the time period just prior to the exam. While a significant interaction
was not found between the intervention and exam performance, the ability of the word count of
writing performed for the intervention to predict exam outcome supports the notion that the
intervention, with appropriate levels of fidelity, could provide a significant interaction between
intervention and outcome. Affording students 10 minutes of time, or potentially more, to write
before an exam is a practice that should be considered by students, teachers, support personnel,
and administrators with a mind toward academic success, retention, and wellness.

Recommendations for Further Research

Findings from the current study imply suggestions for at least two distinct lines of
research: one that centers on test anxiety and a second that centers on central executive function
in attentional control and DMN.

Recommendations for Research in Test Anxiety. With regard to the literature
surrounding test anxiety, writing about worries before an exam as an intervention for test anxiety
warrants further investigation. Considering the asymmetry of the samples, differences in exams
between classes administered subsequent to the intervention, and the unavailability of inter-rater
reliability for exam scores in Sample 2, it is advisable to consider replicating the study design in
a setting where all students are taking the same forced-choice multiple choice exam in same
academic context to eliminate any potential issues with subjectivity associated with scoring
open-ended responses or inter-rater reliability from multiple graders. Further, a larger sample
size in future studies would lend itself to the statistical power of more sophisticated analyses,
specifically when analyses deal strictly with condition differences. With these conditions met, it
WRITING ABOUT WORRIES 70

makes sense to re-examine the questions of whether writing before an exam improves
performance and/or reduces anxiety.

Beyond coming closer to a more standardized context to administer the intervention and
a larger sample to administer it to, manipulating the prompt or the time available to write in such
a way that allows for a more fine-grained analysis of the relationship between the volume of the
writing and the effect of the intervention is advisable. Exploring this relationship is especially
relevant given that one of the primary findings of the current study is that word count from the
intervention predicted exam score, which itself raises interesting questions: Would this
relationship hold in a more standardized context with a larger sample size? If the effect is
replicable, is there a certain volume of writing at which the effect weakens or reverses?

Another line of inquiry relevant to research in test anxiety is if the content of the writing
has any effect on the outcome of the intervention. Simple preliminary analyses lend themselves
to a number of questions: What might account for written expressions of greater anxiety
amongst participants that reported on lower levels of anxiety on surveys? What might account
for written expressions of indifference among participants that reported higher levels of anxiety
on surveys? Has the relationship between self-report survey scores for anxiety and physiological
markers like galvanic skin response or salivary cortisol been previously examined, and if so
might the findings have implications for the findings of the current study? Beyond these
questions, others about the association between categories of iterations captured in the writing
and effects on performance and anxiety offer other avenues for inquiry. Can gross word count be
subdivided into types of thinking that can be related to desirable effects? Are other writing
prompts more conducive to kinds of induction that would optimize performance and anxiety
outcomes? Is writing itself even necessary or would thinking about worries produce the same
WRITING ABOUT WORRIES 71

effects? How might monolingualism and bi- or multilingualism effect outcomes? Are there
differences in outcomes for students who perform the study in their first or in subsequently
learned languages? Could thoughts about other topics or other types of thought produce similar
outcomes to writing? It is highly advisable that further research examine these questions.

Recommendations for research in Attentional Control and Default Mode Network

(DMN). Another line of research has examined the relationship between the different networks
the brain engages when resting, introspecting, or interospecting on the one hand and when
engaging in more active goal-directed tasks that entail attending to external stimuli. This line of
research supports the notion that developing skills conducive to enabling reflection during lapses
in outward attention and allowing for opportunities during which the mind may be allowed to
wander freely may be critical to healthy development and learning in the longer term
(Immordino-Yang, Christodoulou, & Singh, 2012). The kinds of constructive internal reflection
for which this line of research advocates in order to approach an effective balance between
external attention and internal reflection are present in the current study design, but also in others
intended to leverage the neural correlates and cognitive architecture of the central executive
system in ways that facilitate optimal co-regulation of inward and outward looking networks.

Recent work found that engaging in a mindfulness exercise, a kind of focused breathing,
reduced DMN activity relative to undirected mind wandering (Dickenson, Berkman, Arch,
Lieberman, 2013). This raises the question of what might be the most effective way to induce the
mechanism at work in writing about worries when it exerts an effect on anxiety level or performance.
Might effects on performance or anxiety be more pronounced if the mind was allowed to wander
freely rather than to write about worries, or if the intervention was based more on focused
interoception—for example, focusing on the sensation of one’s breath—rather than
WRITING ABOUT WORRIES 72

on reflection, as is the case with writing about worries? Similar questions arise when the focus
of the examination extends beyond immediate effects.

Findings from Dickenson et al. (2013) also suggest that the mindfulness induction they
employed appears be less effortful for experts than for novices, suggesting that it involves skills
which can be developed overtime. This extends work by Taylor et al. (2011) that found
deactivation in DMN among expert meditators when viewing photographs in a mindful way,
suggesting a more active process for novices. Dickinson et al. (2013) found that a brief
introduction and very simple training was sufficient to depress DMN activation. Taken together,
this suggests that the more experienced a meditator, the less effortful DMN deactivation
induction should be. At the same time, DMN deactivation appears to be intrinsic to the task of
mindful breathing such that although it might initially be more effortful, novices with very little
training or experience could reap similar benefits.

Future research would do well to examine DMN deactivation inductions in educational
contexts through analysis of performance and anxiety level. Given that the quantity of activation
has been found to be less important than functional coordination between DMN regions (van den
Heuevel, Stam, Kahn, Hulshoff Pol, 2009), future research might look at the relationship
between DMN depression inductions and other inductions that, rather than simply depress DMN
activity, seek to develop DMN connectivity at rest. The educational implications of an induction,
or inductions, that were able to depress DMN activity in the short-term, allowing students to
reduce deleterious effects of test anxiety, and also developed DMN connectivity over the long
term, which has been associated with intellectual performance (van den Huevel et al., 2009),
would be immense.
WRITING ABOUT WORRIES 73

In summary, data analysis disconfirmed the hypotheses that the intervention would
significantly reduce state anxiety and significantly increase performance, however there were a
number of factors outside of the intervention that lend themselves to an explanation. It stands to
reason that processing efficiency was affected to a degree that did not exert a significant
influence on processing effectiveness. Further, the positive relationship discovered between
word count and performance might indicate issues with fidelity to the treatment among some
participants. Additionally, interactions between preintervention levels of state anxiety and
cognitive test anxiety seem raise interesting questions about the sensitivity of the analyses and
whether there are specific needs best met by the intervention. Finally, explicit instruction in
compensatory strategies may also have contributed to negative findings.

Conclusion

Decrements in performance have long been attributed to test anxiety. A great deal of
literature examines the causes, symptoms, and effects of test anxiety, but relatively little work
offers empirical data from test anxiety interventions. Both Attentional Control Theory and
insights from recent work dealing with the DMN suggest that an optimal balance in co-regulated
inward-looking and outward-looking networks could decrease anxiety and improve performance.
As such, the current study invited students with an impending exam to write about their worries
regarding the exam immediately before taking the exam. Hypotheses about decreases in state
anxiety and improvements in performance were not supported, but additional analysis of the data
showed that word count of writing performed significantly predicted outcome and that
preintervention state anxiety level and cognitive test anxiety level both significantly predicted the
decrease in postintervention state anxiety. It is suggested, then, that future research continue to
examine the usefulness of the intervention, especially for those with high levels of cognitive test
WRITING ABOUT WORRIES 74

anxiety or high levels preexamination state anxiety with a focus on producing a more robust

word count.
WRITING ABOUT WORRIES 75

References

Aronson, J., Fried, C., & Good, C. (2002). Reducing the effects of stereotype threat on
African American college students by shaping theories of intelligence. Journal of
Experimental Social Psychology, 38(2), 113-125.

Ansari, T.L., Derakshan, N. and Richards, A. (2008). Effects of anxiety on task-switching:
Evidence from the mixed antisaccade task. Cognitive, Affective, and Behavioural
Neuroscience, 8(3), 229-238.

Aketkin, M., Karaman, T., Yitiger Senol Y., Erdem S., Erengin H & Akaydin M., (2001).
Anxiety, depression, and stressful life events among medical students: A
prospective study in Antalya, Turkey. Medical Education, 35(1), 12-17.

Atkinson, R.C., & Shiffrin, R.M. (1968). Human memory: A proposed system and its control
processes. In K.W. Spence & J.T. Spence (Eds.), The psychology of learning and
motivation: Advances in research and theory (Vol. 2, pp. 89–195). New York:
Academic Press.

Baddeley, A.D. & Hitch, G. (1974). Working memory. Oxford: Oxford University
press. Baddeley, A.D. (1986). Working memory. Oxford, UK: Clarendon Press.

Baddeley, A.D. (2001). Is working memory still working? American Psychologist, 56(11), 851-
864.

Baddeley, A.D. (2010). Working memory. Current Biology, 20(4), R136-R140.

Beck, A. T., Emery, G., & Greenberg, R. (1985). Anxiety disorders and phobias: A cognitive
perspective. New York: Basic Books.

Beggs, C., Shields, D., Janiszewski Goodin, H. (2011). Using guided reflection to reduce test
anxiety in nursing students. Journal of Holistic Nursing, 29(2), 140-47.
WRITING ABOUT WORRIES 76

Benjamin, M., McKeachie, W. J., Lin, Y., & Holinger, D. P. (1981). Test anxiety: deficits
in information processing. Journal of Educational Psychology, 73(6), 816-824.

Beidel, D. C. & Turner, S. M. (1988). Comorbidity of test anxiety and other anxiety disorders
in children. Journal of Abnormal Child Psychology, 16(3), 275-278.

Bishop, S. J. (2007). Neurocognitive mechanisms of anxiety: An integrative account. Trends in
Cognitive Sciences, 11(7), 307-316.

Bishop, S. J. (2009). Trait anxiety and impoverished prefrontal control of attention. Nature
Neuroscience, 12(1), 92-98.

Bjork, R. A. & Witten, W. B. (1974). Recency-sensitive retrieval processes in long-term free
recall. Cognitive Psychology, 6(2), 173-189.

Bonaccio, S., Reeve, C.L., & Winford, E.C. (2012). Test anxiety on cognitive ability tests can
result in differential predictive validity of academic performance. Personality and
Individual Differences, 52(4), 497-502.

Borkovec, T. D. & Inz, J. (1990). The nature of worry in generalized anxiety disorder: A
predominance of thought activity. Behavior, Research and Therapy, 28(2), 153-158.

Borkovec, T. D. & Roemer, L. (1995). Perceived functions of worry among generalized
anxiety disorder subjects: Distraction from more emotionally distressing topics?
Journal of Behavior Therapy & Experimental Psychiatry, 26(1), 25-30.

Cassady, J. C. & Johnson, R. E. (2002). Cognitive test anxiety and academic performance.

Contemporary Educational Psychology, 27(2), 270-295.

Chapell, M. S., Blanding, Z. B., Silverstein, M. E., Takahashi, M., Newman, B., Gubi,
A., McCann, N. (2005). Test anxiety and academic performance in undergraduate
and graduate students. Journal of Educational Psychology, 97(2), 268-384.
WRITING ABOUT WORRIES 77

Cheung, R. Y. & Au, T. K. (2011.) Nursing students’ anxiety and clinical performance. Journal
of Nursing Education, 50(5), 286-89.

Cizek, G. J. & Burg, S. S. (2006). Addressing test anxiety in a high-stakes enviornment.

Thousand Oaks, CA: Corwin.

Cohen, G.L., Garcia, J., Apfel, N., & Master, A. (2006). Reducing the racial achievement gap:
A social-psychological intervention. Science, 313(5791), 1307-1310.

Cohen, G.L., Garcia, J., Purdie-Vaughns, V., Apfel, N., & Brzustoski, P. (2009). Recursive
processes in self-affirmation: Intervening to close to the minority achievement gap.

Science, 324(5925), 400-403.

Costello, C. G., & Comrey, A. L. (1967). Scales for measuring depression and anxiety. Journal
of Psychology, 66(2), 303-313.

Covington, M. V. (1985). Anatomy of failure-induced anxiety: The role of cognitive mediators.
In R. Schwarzer (Ed.), Self-related cognitions in anxiety and motivation. Hilsdale, NJ:
Lawrence Erlbaum Associates.

De Jong, R., Berendsen, E., Cools, R. (1999). Goal neglect and inhibitory limitations: dissociable
causes of interference effects in conflict situation. Acta Pscyhologia, 101(2-3), 379-394.

Deffenbacher, J. L. (1980).Worry and emotionality in test anxiety. In I. G. Sarason (Ed.), Test
anxiety: Theory, research, and applications (pp. 111-124). Hillsdale, NJ: Erlbaum.

Derakshan, N., & Eysenck, M. W. (2009). Anxiety, processing efficiency, and cognitive
performance: New developments from Attentional Control Theory. European
Psychologist, 14(2), 168-176.

Derakshan, N., Ansari, T. L., Hansard., M. Shoker, L., Eysenck, M. W. (2009). Anxiety,
inhibition, efficiency, and effectiveness. Experimental Psychology, 56(91), 48-55.
WRITING ABOUT WORRIES 78
Derakshan, N., Smith, S. Eysenck, M. W. Effects of state anxiety on performance using a task-
switching paradigm: An investigation of Attentional Control Theory. Psychonomic
Bulletin & Review, 16(6), 1112-1117.
Dickenson, J., Berkman, E. T., Arch, J., Lieberman, M. D. (2013). Neural correlates of focused
attention during a brief mindfulness induction. SCAN, 8(1), 40-47.

Drevets, W. C., (2001). Neuroimaging and neuropathological studies of depression: implication
for the cognitive-emotional features of mood disorders. Current opinion in Neurobiology,
11(2), 240-249.

Dutke, S. & Stober, J. (2001). Test anxiety, working memory, and cognitive performance:
Supportive effects of sequential demands. Cognition and Emotion, 15(3), 381-389.

Ergene, T. (2003). Effective interventions on test anxiety reductions: A meta-analysis. School
Psychology International, 24(3), 313-328.

Everson, H. T., Millsap, R. E., Rodriguez, C. M. (1991) Isolating gender differences in test
anxiety: a confirmatory factor analysis of the test anxiety inventory. Educational and
Psychological Measurement, 51(1), 243-251.

Eysenck, M.W., & Calvo, M. G. (1992). Anxiety and performance. The processing efficiency
theory. Cognition and Emotion, 6(6), 409-434.

Eysenk, M.W., Payne, S., & Derakshan, N. (2010). Trait anxiety, visuospatial processing,
and working memory. Cognition and Emotion 19(8), 1214-1228.

Eysenck, H. J. (1979). The conditioning model of neurosis. Behavioral and Brain Sciences,
2(2), 155-199.

Eysenck, M. W., Derakshan, N., Santos, R., & Calvo, M. G. (2007). Anxiety and
cognitive performance: Attentional control theory. Emotion, 7(2), 336-353.
WRITING ABOUT WORRIES 79

Fales, C.L., Barch, D.M., Burgess, G.C., Shaefer, A., Mennin, D.S., Gray, J.R., & Braver, T.S.
(2008). Anxiety and cognitive efficiency: Differential modulation of transient and
sustained neural activity during a working memory task. Cognitive, Affective, &
Behavioral Neuroscience, 8(3), 239-253.

Feldman, L. A. (1993). Distinguishing depression and anxiety in self-report: Evidence from
confirmatory factor analysis on nonclinical and clinical samples. Journal of
Consulting and Clinical Psychology, 61(4), 631-638.

Finkelstein, C., Brownstein, A., Scott, C., Lan, Y. (2007). Anxiety and stress reduction in
medical education. Medical Education, 41(3), 258-264.

Geller, B. (2011, Februarly 14). Afraid of taking that test? Write about it. Journal Watch
Psychiatry.

Golden, J. S., & Rosen, A. C. (1975). A group dynamics course for medical students.

International Journal for Group Psychotherapy, 25(1/4), 305-313.

Gorsuch, R. L. (1966). The general factor in the test anxiety questionnaire. Psychological
reports, 19(1), 308.

Gotlib, I. H. (1984). Depression and general psychopathy in university students. Journal of
Abnormal Psychology, 93(1), 19-30.

Graham, S., Vettarino, A. N., Seifelden, R., Singal, B. (2010). A trial of virtual hypnosis to
reduce stress and test anxiety in family medicine [Editorial]. Family Medicine, 42(2),
85-86.

Grecius, M. D., Krasnow, B. Reiss, A. L., Menon, V. (2003). Functional connectivity in the
resting brain: A network analysis of the default mode hypothesis. Proceedings of
the National Academy of Sciences, 100(1), 253-258.
WRITING ABOUT WORRIES 80

Gusnard, D. A., & Raichle, M. E. (2001). Searching for a baseline: Functional imaging and the
resting human brain. Nature Reviews Neuroscience, 2(10), 685-694.

Hembree, R. (1988). Correlates, causes, effects, and treatment of test anxiety. Review
of Educational Research, 58(1), 47-77.

Hill, K. T., & Wigfield, A. (1984). Test anxiety: A major educational problem and what can
be done about it. The Elementary School Journal, 85(1), 106-126.

Holtzworth-Munroe, A., Munroe, M. S., Smith, R. E. (1985). Effects of a stress management
training program on first- and second-year medical students. Academic Medicine, 60(5),
47-77.

Hosanagar, A., Mattai, A., Lalonde, F., Greenstein, D., Clasen, L.,…Gogtay, N. (2010).

Thalamic development in childhood onset schizophrenia: A volumetric analysis.

Schizophrenia Research, 117(2/3), 323.

Hulleman, C.S., & Harackiewicz, J.M. (2009). Making education relevant: Increasing interest
and performance in high school science classes. Science, 326(5958), 1410-1412.

Ikeda, M., Iwanaga, M., Seiwa, H. (1996). Test anxiety and working memory system. Perceptual
and Motor Skills, 82(3), 1223-1231.

Immordino-Yang, M.H., Christodoulou, J.A., Sing, V. (2012). Rest is not idleness: implications
of the brain's default mode for human development and education. Perspectives on
Psychological Science, 7(4), 352-364.

Jamieson, J. P., Nock, M. K., Mendes, W. B., (2009). A metacognitive perspective on the
cognitive deficits experienced in intellectually threatening enviornments. Personality and
Social Psychology Bulletin, 35(5), 584-596.

Kelly, J. A., Bradlyn, A. S., Dubbert, P. M., St. Lawrence, J. S. (1982). Stress management
WRITING ABOUT WORRIES 81

training in medical school. Journal of Medical Education, 57(2), 91-99. Kiecolt-
Glaser, J.K., Glaser, R., Strain, E.C., Stout, J.C., Tarr, K. L., Holliday, J. E., Speicher,

C. E. (1984). Modulation of cellular immunity in medical students. Journal of
Behavioral Medicine, 9(1), 5-21.

Kiessling, C., Schubert, B., Scheffner, D. & Burger, W. (2004). First year medical students’
perceptions of stress and support: A comparison between reformed and traditional track
curricula. Medical Education, 38(5), 504-09.

Kirkland, K. Hollandsworth, J. G. (1980). Effective test taking: skills-acquisition versus anxiety-
reduction techniques. Journal of Consulting and Clinical Psychiatry, 48(4), 431-439.

Klein, K., & Boals, A. (2001). Expressive writing can increase working memory capacity.

Journal of Experimental Plsychology, 130(3), 520-533.

Klein, R. G. (2009). Anxiety Disorders. Journal of Child Psychology and Psychiatry, 50(2), 153-
162.

Klose Smith, C., Peterson, D. F., Degenhardt, B. F. & Johnson, J. C. (2007). Depression,
anxiety, and perceived hassles among entering medical students. Psychology, Health &
Medicine, 12(1), 31-9.

Lang, J.W.B., & Lang, J. Practical implications of test anxiety tools. Science, 332(13), 791-792
Lazarus, R. S. (1966). Psychological stress and the coping process. New York: McGraw-Hill.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Liebert, R.M. & Morris, L. W. (1967). Cognitive and emotional components of test anxiety: A

distinction and some initial data. Psychological Reports, 20(3), 975-78.

Maddock, R. J. (1999). The retrosplenial cortex and emotion: New insights from functional
neuroimaging of the brain. Trends in Neurosciences, 22(7), 310-316.
WRITING ABOUT WORRIES 82

Maddock, R. J., Garrett, A. S., & Buonocore, M. H. (2003). Posterior cingulate cortex
activation by emotional words: fMRI evidence from a valence decision task. Human
Brain Mapping, 18(1), 30-41.

Markman, U., Balik, C., Braunstein-Becovitz, H. & Eherenfeld, M. (2011). The effect of
nursing students’ health beliefs on their willingness to seek treatment for test anxiety.
Journal of Nursing Education, 50(5), 249-252.

Marteau, T. M., & Bekker, H. (1992). The development of a six-item short-form of the state
scale of the Spielberger State-Trait Anxiety Inventory (STAI). British Journal of Clinical
Psychology, 31(3), 301-306.

Mattarella-Micke, A., Mateo, J., Foster, K., Beilock, S.L., Kozak, M.N. Choke or Thrive? The
relation between salivary cortisol and math performance depends on individual
differences in working memory and math anxiety. Emotion, 11(4), 1000-1005.

Mazoyer, B., Zago, L., Mellet, E., Bricogne, S., Etard, O., Houde, O., Tzourio-Mazoyer, N.
Cortical networks for working memory and executive fuctions sustain the conscious
resting state in man. Brain Research Bulletin, 54(3), 287-298.

McGuire, D. P., Mitic, W., & Neumann, B. (1987). Perceived stress in adolescents: What normal
teenagers worry about. Canada's Mental Health, 35(2), 2-5.

Mitchell, K. R., & Ng, K. T. (1972). Effects of group therapy on the academic achievement of
test-anxious students. Journal of Counseling Psychology, 19(6), 491-497.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D.
(2000). The unity and diversity of executive functions and their contributions to complex
"frontal lobe" tasks: A latent variable analysis. Cognitive Psychology, 41(1), 49-100.

Miyake, A., Kost-Smoth, L.E., Finkelstein, N.D., Pollock, S.J., Cohen, G.L., & Ito, A. (2010).
WRITING ABOUT WORRIES 83

Reducing the gender achievement gap in college science: A classroom study of values
affirmation. Science, 330(6008), 1234-1237.

Moffat, K. J., Mcconnachie, A., Ross, S., & Morrison, J. M. (2004). First year medical
student stress and coping in a problem-based learning medical curriculum. Medical
Education, 38(5), 482-491.

Morris, L. W., Davis, M. A., & Hutchings, C. H. (1981). Cognitive and emotional components of
anxiety: Literature review and a revised worry-emotionality scale. Journal of Educational
Psychology, 73(4), 541-555.

Mueller, J. H. (1980). Test anxiety and the encoding and retrieval of information. In I.G.
Sarason (Ed.), Test Anxiety: theory, research, and applications(pp.63-86). Hillsdale,
NJ: Erlbaum.

Nathan, R. G., Nixon, F. E., Robinson, L. A., Bairnsfather, L., Allen, J. H., Hack, M. (1987).
Effects of a stress management course on grades and health of first-year students.

Medical Education, 62(6), 514-527.

Naveh-Benjamin, M. (1991). A comparison of training programs intended for different types of
test anxious students: Further support for an information-processing model. Journal of
Educational Psychology, 83(1), 134-139.

Palan, B. M., & Chandwani, S. (1989). Coping with examination stress through hypnosis:
an experimental study. American Journal of Clinical Hypnosis, 31(3), 173-180.

Pani, S. C., Al Askar, A. M., Al Mohrij, S. I., Ohali, T. A., (2011). Evaluation of stress in final-
Saudi dental students using salivary cortisol as a biomarker. Journal of Dental
Education, 75(3), 377.

Pekrun R. (2006). The control-value theory of achievement emotions: Assumptions, corollaries,
WRITING ABOUT WORRIES 84

and implications for educational research and practice. Educational Psychology Review,

18(4), 315-341.

Pekrun, R., Goetz, T., Frenzel, A. C., Barchfield, P., & Perry, R. A. (2011). Measuring emotions
in students’ learning and performance: The achievement emotions questionnaire (AEQ).

Contemporary Educational Psychology, 36(1), 36-48.

Pekrun, R. Goetz, T., Titz, W., & Perry, Raymond P. (2002). Academic emotions in students’
self-regulated learning and achievement: A program of qualitative and quantitative
research. Educational Psychologist, 37(2), 91-105.

Pennebaker, J. W. (1984). Accuracy of physical symptoms. In A. Baum, S. Taylor, & J.
Singer (Eds.), Handbook of psychology and helath, Volume 4 (pp.189-217). Hillsdale,
NJ: Erlbaum.

Peterlini, M., Tiberio, I.F.L.C., Saadeh, A., Pereira, J. C. R. & Martins, M. A. (2006). Anxiety
and depression in the first year of medical residency training. Medical Education, 36(1),
66-72.

Ping, L.T., Subramaniam, K., Krishnaswamy, S. (2008). Test Anxiety: State, trait, and
relationship with exam satisfaction. Malaysian Journal of Medical Sciences, 15(2), 18-
23.

Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain. Annual Review
of Neuroscience, 13(1), 25-42.

Posner, M. I., & Rothbart, M. K. (2005). Influencing brain networks: Implications for education.

Trends in Cognitive Sciences, 9(3), 99-100.

Powell, D. H. (2004). Behavioral treatment of Debilitating Test Anxiety among medical
students. Journal of Clinical Psychology, 60(8), 853-865.
WRITING ABOUT WORRIES 85

Schmader, T., Forbes, C. E., Zhang, S., Mendes, W.B. (2009). A metacognitive perspective on
the cognitive deficits experienced in intellectually threatening environments.
Personality and Social Psychology Bulletin, 35(5), 584-58.

Tonneatto, T. & Nguyen, L. (2007). Does mindfulness meditation improve anxiety and mood
symptoms? A review of the controlled research. Canadian Journal of Psychiatry, 52(4),
260-66.

Reeve, C.L., & Bonnaccio, S. Does test anxiety induce measurement bias in cognitive
ability tests? Intelligence, 36(6), 526-538

Reteguiz, J. (2006). Relationship between anxiety and standardized patient test performance
in the medicine clerkship. Journal of General Internal Medicine, 21(5), 415-18.

Righi, S., Mecacci, L., Viggiano, M. P. (2009). Anxiety, cognitive self-evaluation, and
performance: ERP correlates. Journal of Anxiety Disorders, 23(8), 1132-1138.

Rozenweig, S., Reibel, D. K., Greeson, J. M., Brainard, G. C., Hojat, M. (2003). Mindfulness-
based stress reduction lowers psychological distress in medical students. Teaching and
Learning in Medicine, 15(2), 88-92.

Sanisgiry, S.S. & Sail, K. (2006). Effect of students’ perceptions of course load on test anxiety.

American Journal of Pharmaceutical Education, 70(2), Article 26.

Sarason, S. B., & Mandler, G. (1952). A study of anxiety and learning. Journal of Abnormal
Social Psychology, 47(2), 166-173.

Sarason, I. G. (1984). Stress, anxiety, and cognitive interference: Reactions to tests. Journal of
Personality and Social Psychology, 46(4), 929-938.

Sarason, S. B., Davison, K. S., Waite, R. R., Lighthall, F. K., & Ruebush, B. K. (1960). Anxiety
in elementary school children: A report of research. New York: Wiley.
WRITING ABOUT WORRIES 86

Sarikaya, O., Civaner, M. & Kalaca, S. (2006). The anxieties of medical students related
to clinical training. International Journal of Clinical Practice, 60(11), 1414-18.

Sassenrath, J. M. (1964). A factor analysis of rating-scale items on the test anxiety questionnaire.

Journal of Consulting Psychology, 28(4), 371-377.

Sassenrath, J. M., Knight, H. R., Kaiser, H. F. (1965). Relating factors from anxiety scales
between two samples. Psychological Reports, 17(2), 407-416.

Savostyanov, A. N., Tsai, A. C., Liou, M., Levin, E. A., Lee., J-D., Yurganov, A. V., &
Knyazev, G. G. (2009). EEG-correlates of trait anxiety in the stop-signal paradigm.

Neuroscience Letters, 44(2), 112-116

Schmader, T., & Johns, M. (2003). Converging evidence that stereotype threat reduces working
memory capacity. Journal of personality and social psychology, 85(3), 440-452.

Schutz, P. A., & Davis, H. A. (2000). Emotions and self-regulation during test taking.

Educational Psychologist, 35(4), 243-256.

Schulman, G. L., Fiez, J. A., Corbetta, M. Buckner, R. L. Miezin, F. M., Raichele, M. E., &
Petersen, S. E. (1997). Common blood flow changes across visual tasks: Decreases
in cerebral cortex. Journal of Cognitive Neurosciences, 9(5), 648-663.

Schwarzer, R. (1984). Worry and emotionality as separate components in test anxiety.

International Review of Applied Psychology, 33(2), 205-220.

Serovich, J. M., Kimberly, J. A., Mosack, K. E., & Lewis, T. L. (2001). The role of family
and friend social support in reducing emotional distress among HIV-positive women.
AIDS care, 13(1), 335-341.

Shapiro, S.L., Astin, J.A., Bishop, S.R., Cordova, M. (2005). Mindfulness-based stress reduction
WRITING ABOUT WORRIES 87

for health care professionals: Results from a randomized trial. International Journal of
Stress Management, 12(2), 164-176.

Shapiro, S. L, Schwartz, G. E., Bonner, G. (1998). Effects of mindfulness-based stress reduction
on medical and premedical students. Journal of Behavioral Medicine, 21(6), 581-599.

Sharif, F. & Armitage, P. (2004). The effect of psychological and educational counseling in
reducing anxiety in nursing students. Journal of Psychiatric and Mental Health Nursing,
11(4), 386-392.

Simpson, J. R., Jr., Drevets, W. C., Snyder, A. Z., Gusnard, D. A., & Raichle, M. E. (2001).
Emotion-induced changes in human medial prefrontal cortex: During anticipatory
anxiety. Proceedings of the National Academy of Sciences, 98(2), 688-693.

Smallwood, J. Fishman, D. J., & Schooler, J. W. (1997). Psychonomic Bulletin & Review, 14(3),
pp. 527-533.

Smith, E. E., & Jonides, J. (1999). Storage and executive processes in the frontal lobes. Science,

283(5408), 1657-1661.

Smith, R. C., Elwood, L. S., Feldner, M. T., & Olatunji, B. O. In D. L. Segal & M.
Hersen (Eds.), Dianostic Interviewing. (pp. 125-151). New York: Springer.

Smyth, J. M. (1998). Written emotional expression: Effect sizes, outcome types, and
moderating variables. Journal of Consulting and Clinical Psychology, 66(1), 174-184.

Spielberger, C. D. (1973). Manual for the state-trait anxiety inventory for children. Palo Alto,
CA: Consulting Psychologists.

Speilberger, C. D. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA:
Consulting Psychologists.

Speilberger, C. D., Gaudry, E., & Vagg, P. (1975). Validation of the state-trait distinction in
WRITING ABOUT WORRIES 88

anxiety research. Multivariate Behavioral Research, 10(3), 331-341.

Speilberger, C. D., Gorsuch, R. L., & Luschene, R. E. (1970). The state-trait anxiety inventory:
Test manual. Palo Alto, CA: Consulting Psychologists.

Stecker, J. (2004). Well-being in an academic environment. Medical Education, 38(5), 465-478.
Stevens, C., Lauinger, Brittni; & Neville, H. (2009). Differences in the neural mechanisms of

selective attention in children from different socioeconomic backgrounds: An event-
related brain potential study. Developmental Science, 12(4), 634-646.

Sub, A., & Prabha, C. (2003). Academic performance in relation to perfectionism, test
procrastination, and test anxiety of high school children. Psychological Studies, 48(1), 7-
81.

Taylor, V.A., Grant, J. Deneault, V., Scavone, G., Breton, E., Roffe-Vidal, S….Beauregard, M.

(2011). Impact of mindfulness on the neural responses to emotional pictures in
experienced and beginner meditators. Neuroimage, 57(4), 1524-1533.

Tobias, S. (1979). Anxiety research in educational psychology. Journal of
Educational Psychology, 53,573-582.

Tonneatto, T. & Nguyen, L. (2007). Does mindfulness meditation improve anxiety and mood
symptoms? A review of the controlled research. Canadian Journal of Psychiatry, 52(4),
260-66.

van den Huevel, M. P., Stam, C. J., Kahn, R. S., & Hulshoff Pol, H. E. (2009). Efficiency of
functional brain networks and intellectual performance. Journal of Neuroscience, 29(23),
7619-7624.

Vasegh, S. & Mohammadi, MR. (2007). Religiosity, anxiety, and depression among a sample of
WRITING ABOUT WORRIES 89

Iranian medical students. International Journal of Psychiatry in Medicine, 37(2), 213-
237.

Walton, G.M., & Cohen, G. L. (2007). A question of belonging: Race, social fit, and
achievement. Journal of Personality and Social Psychology, 92(1), 82-96.

Walton, G.M., & Cohen, G. L. (2011). A brief social-belonging intervention improves academic
and health outcomes among minority students. Science, 331(6023), 1447-1451.

Wegner, D. M. Schneider, D. J., Carter, 3rd, S. R., & White, T. L. (1987). Paradoxical effects
of thought suppression. Journal of Personality and Social Psychology, 53(1), 5-13.

Whitehouse, W. G., Bloom, P. B., Zaugg, L; Orne, M. T., Dinges, D. F., Orne, E. C….Carlin,
M. M. (1996). Psychosocial and immune effects of self-hypnosis training for stress
management throughout the first semester of medical school. Psychosomatic Medicine,
58(3), 249-257.

Wine, J. (1971). Test anxiety and direction of attention. Psychological Bulletin, 76(2), 92-104.
Yeager, D.S. & Walton, G.M. (2011). Social-Psychological interventions in education: They’re

not magic. Review of Educational Research, 81(2), 267-301.

Zald, D. H., Mattson, D. L., & Pardo, J. V. (2002). Brain activity in ventromedial prefrontal
cortex correlates with individual differences in negative affect. Proceedings of the
National Academy of Sciences, 99(4), 2450-2454.

Zeidner, M. (1998). Test anxiety: The state of the art. New York: Plenum.
WRITING ABOUT WORRIES 90

APPENDICIES

Appendix 1: Demographics Questionnaire

Demographics Questionnaire

Age:

_____ years old

Gender:

□Male
□Female

Race:

□White □African-
American

□Hispanic/Latino
□Asian

□Native Hawaiian or Pacific Islander
□American Indian or Alaskan Native
□Other, Not Listed. Please Identify ___________________
□Two or more races. Please identify___________________

Student ID number: ______________________________
WRITING ABOUT WORRIES 91

Appendix 2: Cognitive Test Anxiety Scale

Cognitive Test Anxiety Scale (Cassady & Johnson, 2004)

(A = Not at all typical of me, B = Only somewhat typical of me, C = Quite typical of me, and D =
Very typical of me.)

1. I lose sleep over worrying about examinations.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
2. While taking an important examination, I find myself wondering whether the other students
are doing better than I am.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
3. I have less difficulty than the average college student in getting test instructions straight.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
4. I tend to freeze up on things like intelligence tests and final exams.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
5. I am less nervous about tests than the average college student.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
6. During tests, I find myself thinking of the consequences of failing.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
7. At the beginning of a test, I am so nervous that I often can’t think straight.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
WRITING ABOUT WORRIES 92

8. The prospect of taking a test in one of my courses would not cause me to worry.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
9. I am more calm in test situations than the average college student.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
10. I have less difficulty than the average college student in learning assigned chapters in
textbooks.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
11. My mind goes blank when I am pressured for an answer on a test.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
12. During tests, the thought frequently occurs to me that I may not be too bright.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
13. I do well in speed tests in which there are time limits.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
14. During a course examination, I get so nervous that I forget facts I really know.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
15. After taking a test, I feel I could have done better than I actually did.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
16. I worry more about doing well on tests than I should.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
WRITING ABOUT WORRIES 93

17. Before taking a test, I feel confident and relaxed.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
18. While taking a test, I feel confident and relaxed.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
19. During tests, I have the feeling that I am not doing well.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
20. When I take a test that is difficult, I feel defeated before I even start.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
21. Finding unexpected questions on a test causes me to feel challenged rather than panicky.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me

22. I am a poor test taker in the sense that my performance on a test does not show how much I
really know about a topic.

1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
23. I am not good at taking tests.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me

24. When I first get my copy of a test, it takes me a while to calm down to the point where I
can begin to think straight.

1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me

25. I feel under a lot of pressure to get good grades on tests.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
WRITING ABOUT WORRIES 94

26. I do not perform well on tests.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
27. When I take a test, my nervousness causes me to make careless errors.
1 2 3 4
Not at all typical of me Only somewhat typical of me Quite Typical Very Typical of me
WRITING ABOUT WORRIES 95

Appendix 3: Trait Anxiety Scale

Costello-Comrey Anxiety Scale (Costello-Comrey Anxiety Scale, 1967)

1. I get rattled easily.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

2. When faced with excitement or unexpected situations, I become nervous and jumpy.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

3. I am calm and not easily upset

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

4. When things go wrong I get nervous and upset instead of calmly thinking out a solution.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

5. It makes me nervous when I have to wait.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

6. My hand shakes when I try to do something.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always
WRITING ABOUT WORRIES 96

7. I am a tense “high strung” person.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

8. I am more sensitive than most other people.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always

9. I am a very nervous person.

1 2 3 4 5 6 7 8 9

Never—Almost never—Rarely—Occasionally—Fairly often—Frequently—Very Frequently—Almost always—Always
WRITING ABOUT WORRIES 97

Appendix 4: State Anxiety Scale

Marteau and Bekker State-Trait Anxiety Inventory State Scale Short-Form (Marteau and

Bekker 1992)

(1 = Not at all typical of me, 2 = Only somewhat typical of me, 3 = Quite typical of me, and 4 =
Very typical of me.)

1) I feel calm
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
2) I feel tense
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
3) I feel upset
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
4) I am relaxed
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
5) I feel content
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
6) I am worried
1 2 3 4
Strongly Disagree Somewhat Disagree Somewhat Agree Strongly Agree
WRITING ABOUT WORRIES 98

Appendix 5: Prompt for Treatment Group

Prompt for Treatment Group (Ramirez & Beilock, 2011)

This short assignment will not be graded. Please put away all your materials, (i.e., notes,
textbooks, calculators) except for a pencil and do not talk during this time. You and your
classmates are about to start your science examinations. However, before beginning the exams,
everyone will take the next 10 minutes to complete a short exercise related to the exam they are
about to take. During this exercise, some students will be asked to just think about the upcoming
exam and other students will be asked to put their thoughts down in writing. Every student is
being asked to do something a little different.

We would like YOU to take the next 10 minutes to write as openly as possible about your
thoughts and feelings regarding the exam you are about to take. In your writing, I want you to
really let yourself go and explore your emotions and thoughts as you are getting ready to start the
exam. You might relate your current thoughts to the way you have felt during other similar
situations at school or in other situations in your life. Please try to be as open as possible as you
write about your thoughts at this time.

There will be no identifying information on your essay. None of the teachers can link
your writing or any other information to you. If you finish early, please just sit quietly and wait
for the teacher’s instructions. You may end up sitting quietly for several minutes while your
classmates finish the tasks they were asked to do. That’s ok. You will be given plenty of time to
complete the upcoming exam. This task will only take about 10 minutes in total. Please begin.
WRITING ABOUT WORRIES 99

Appendix 6: Prompt for Control Group

Prompt for Control Group (Ramirez & Beilock, 2011)

This short assignment will not be graded. Please put away all your materials (i.e., notes,
textbooks, calculators) except for a pencil and do not talk during this time. You and your
classmates are about to start your science examinations. However, before beginning the exams,
everyone will take the next 10 minutes to complete a short exercise related to the exam they are
about to take. During this exercise, some students will be asked to just think about the upcoming
exam and other students will be asked to put their thoughts down in writing. Every student is
being asked to do something a little different.

We would like YOU to take the next 10 minutes to sit quietly and think about one topic
that you feel will NOT be covered on the exam you are about to take. Think about various
reasons why this topic might not be covered in a very factual manner (e.g., it is not my teacher’s
favorite topic, we spent a short time on it, etc.). Once you have done this, please just sit quietly
for several minutes while your classmates finish the tasks they were asked to do. That’s ok. You
will be given plenty of time to complete the upcoming exam. This task will only take about 10
minutes in total. Please begin.

Abstract (if available)

Abstract The phenomenon of test anxiety has long been associated with decrements in performance (Zeidner, 1998) and has been found to affect up to 40% of all students (Cizek & Berg, 2006). Because a substantial number of students that deal with test anxiety perform below their ability on exams, test anxiety during cognitive ability tests has also been identified as a root cause of differential predictive validity of academic performance (Bonnaccio, Reeve & Winford, 2011). In addition, recent developments in cognitive psychology and neuroscience have led to a greater understanding of the neurological and psychological mechanisms at work in test anxiety (e.g., Eysenck & Derakshan, 2011

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Individual differences in heart rate response and expressive behavior during social emotions: effects of resting cardiac vagal tone and culture, and relation to the default mode network

PDF

Asynchronous online education and introductory statistics: The role of learner characteristics

PDF

Motivational and academic outcomes in retained middle school students

PDF

Asynchronous online education and introductory statistics: The role of learner characteristics

PDF

The examination of academic self-regulation, academic help-seeking, academic self-efficacy, and student satisfaction of higher education students taking on-campus and online course formats

PDF

Motives and methods: motivation, learning approaches, and academic achievement of students during first year transition to medical school

PDF

Beliefs and behaviors of online and face-to-face students: evaluating differences in collaboration, help‐seeking, and ethics

PDF

A marriage and family therapy trainee curriculum: college student success in academic self-regulation

PDF

Community college students in STEM: a quantitative study investigating the academic beliefs of students enrolled in online and on campus information technology courses

PDF

Student academic self‐efficacy, help seeking and goal orientation beliefs and behaviors in distance education and on-campus community college sociology courses

PDF

Factors associated with second language anxiety in adolescents from different cultural backgrounds

PDF

Service provider compliance with regional center service agreement: a gap analysis

PDF

Parental involvement and student motivation: A quantitative study of the relationship between student goal orientation and student perceptions of parental involvement among 5th grade students

PDF

Age differences in the effects of anxiety on selective attention: a test of the vigilance-avoidance hypothesis

PDF

Academic beliefs and behaviors in on-campus and online general education biology classes

PDF

The transfer of online instruction to TESOL candidates' perceived self-efficacy of teaching English language learners

PDF

A comparison of student motivation by program delivery method: self-efficacy, goal orientation, and belongingness in a synchronous online and traditional face-to-face environment

PDF

Violence experienced by registered nurses working in hospitals: an evaluation study

PDF

Supporting learners with attention deficit hyperactivity disorder: a professional development curriculum for elementary teachers

PDF

Leadership readiness: evaluating the effectiveness for developing managers as coaches

Asset Metadata

Creator Blank-Spadoni, Nicholas (author)

Core Title Writing about worries as an intervention for test anxiety in undergraduates

School Rossier School of Education

Degree Doctor of Education

Degree Program Education

Publication Date 07/10/2013

Defense Date 05/01/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag academic performance,attentional networks,default mode network,educational psychology,expressive writing,Neuroscience,OAI-PMH Harvest,test anxiety

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hirabayashi, Kimberly (committee chair), Immordino-Yang, Mary Helen (committee member), Keim, Robert G. (committee member), Seli, Helena (committee member)

Creator Email nsblankspadoni@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-285650

Unique identifier UC11294022

Identifier etd-BlankSpado-1750.pdf (filename),usctheses-c3-285650 (legacy record id)

Legacy Identifier etd-BlankSpado-1750.pdf

Dmrecord 285650

Document Type Dissertation

Format application/pdf (imt)

Rights Blank-Spadoni, Nicholas

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA