University of Southern California Dissertations and Theses

Affect, digital media use, physical activity, and ADHD in youth

(USC Thesis Other)

Affect, digital media use, physical activity, and ADHD in youth

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Affect, Digital Media Use, Physical Activity, and ADHD in Youth
By
Chaelin Karen Ra, M.P.H.
_____________________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
in PREVENTIVE MEDICINE

Institute for Health Promotion and Disease Prevention Research
Department of Preventive Medicine
University of Southern California

May, 2019

Ra, Dissertation
1

1
Acknowledgements

I would like to thank my dissertation committee, Drs. Steven Yale Sussman, Jennifer Unger, and
Erik Schott for great support throughout the entire dissertation process and Drs. Chih Ping Chou,
Adam Leventhal, Genevieve Dunton, and Jimi Huh throughout graduation school. I have learned
valuable lessons from all of you, and I do appreciate all the advice given and opportunities to
work with each of you individually. Now I see that I have grown through my experiences during
the doctoral training. I especially wanted to thank Dr. Sussman for his support and giving me
sincere guidance as I have completed my doctoral studies.

I would also like to thank my Marny Barovich for everything she has done for me. I truly believe
that I would not have made it this far without her. There were many moments that I felt hopeless
and did not know what to do. You always were there and guided me towards the right direction.
You are awesome and very important to me! I thank Sherri Fagan – you could not imagine how
much I was touched and encouraged by your words.

I would like to thank all of my Soto and HBR friends. I never imagined that I would meet such
wonderful friends like you and work in such a welcoming and supportive environment before I
came here. I feel so lucky to be friends with you and really thankful for all of your advice,
support, and encouragement. I hope to stay in touch with all of you as we move ahead in our
careers and find ourselves on different paths. I am very excited for what the future holds for us
all!

Lastly, huge thanks to my family and my little sister, Nazife Pehlivan, who have believed in me
and provided endless support throughout the years.

Ra, Dissertation
2

2
Table of Contents

ABSTRACT..................................................................................................................... 7
SPECIFIC AIMS.............................................................................................................. 8
CHAPTER 1: BACKGROUND AND SIGNIFICANCE
Prevalence of ADHD………………...……………………………………………….………… 12
Definition of Affect………………...…………………………………………………………… 12
Dimensions of Affect States: Positive Affect and Negative Affect…………………..………… 13
Positive and Negative Urgency as Affect Traits…….…………………………..……………… 14
Underlying Mechanisms between Affective States and Positive and Negative Urgency……… 14
Possible Mechanisms Linking Affect with ADHD………………………………………. 15
Possible mechanisms Digital Media Use and ADHD…………….…...…….......……………… 16
Underlying Mechanisms between ADHD and Physical Activity……………………….……… 18
Current Research Gaps……………….……………………..………………………………..… 19
Overview of dissertation studies…………………………...…………………………….…...… 20
CHAPTER 2: EXPLORING THE ASSOCIATION OF AFFECTIVE STATES AMONG
CHILDREN AND THEIR SYMPTOM STATUS
INTRODUCTION………………………………………………………………………… 22
The mean level and variability of affective states in mental health research………………… 23
Lack of Consensus in previous studies…………………………………………………………. 24
Variability of Affect as a possible moderator………………………..…………………………. 25
Current Research Gaps………………………..……………………………………………...… 26
Specific Aims and Hypotheses…………………………….....……..………………………... 26
METHODS………………………………………………………………………………… 28
Data, Participants and Procedures………………….………………….………………….… 28
Measures……………….……………………….………………...….…………………….… 29
Statistical Analysis……………………………………………………………………….….…. 33
RESULTS…………………………….……………………………....……………………. 34
DISCUSSION…………………………….………………………………………..………. 40
Limitations………………………..……….…………….……………………………….…… 41
Implications………………………..……….…………….……………………………….…… 42
Future Research…...………………..……….…………….……………………………….…… 44
CONCLUSION………………………..…….…………….……………………………….…… 44
CHAPTER 3: ASSESSING THE ASSOCIATION BETWEEN DIGITAL MEDIA USE
AND ADHD SYMPTOM STATUS AND THE MODERATING EFFECTS OF
AFFECTIVE TRAITS BETWEEN THE TWO AMONG ADOLESCENTS
INTRODUCTION………………………………………………………………………… 45
Prevalence of Digital Media Use of Adolescents………………………………………….… 45
Possible Mechanisms Linking Digital Media Use with ADHD ………… ………… … ………… 46
Digital Media Use and Urgency………………………………………………………………. 46
Ra, Dissertation
3

3
Specific Aims and Hypotheses…………………………….....……..………………………... 48
METHODS………………………………………………………………………………… 50
Data, Participants and Procedures………………….………………….………………….… 50
Measures……………….……………………….………………...….…………………….… 53
Statistical Analysis……………………………………………………………………….….…. 56
RESULTS…………………………….……………………………....……………………. 57
DISCUSSION…………………………….………………………………………..………. 68
Limitations………………………..……….…………….……………………………….…… 70
Implications………………………..……….…………….……………………………….…… 71
Future Research…...………………..……….…………….……………………………….…… 72
CONCLUSION………………………..…….…………….……………………………….…… 72
CHAPTER 4: EXAMINING THE MODERATING EFFECTS OF PHYSICAL
ACTIVITY IN THE ASSOCIATION BETWEEN AFFECTIVE STATES AND ADHD
AND BETWEEN DIGITAL MEDIA USE AND ADHD
INTRODUCTION………………………………………………………...………………… 73
Physical activity and ADHD……………………………………………………...……..……… 73
Physical Activity as a potential moderator between Affect and ADHD………………...… 73
Physical Activity as a potential moderator between Digital Media Use and ADHD………… 74
Specific Aims and Hypotheses…………………………….....……..………………………... 75
METHODS………………………………………………………………………………… 77
Data, Participants and Procedures………………….………………….………………….… 77
Measures……………….……………………….………………...….…………………….… 80
Statistical Analysis……………………………………………………………………….….…. 85
RESULTS…………………………….……………………………....……………………. 87
DISCUSSION…………………………….………………………………………..………. 94
Limitations………………………..……….…………….……………………………….…… 96
Implications………………………..……….…………….……………………………….…… 96
CONCLUSION………………………..…….…………….……………………………….…… 97
CHAPTER 5: OVERALL DISCUSSION & CONCLUSIONS
Summary of findings……………………………………………...………...………………… 98
Theoretical Implications…………………….……………………….………...……..……… 99
Methodological Implications …………….………………………………...…………… 100
Political Implications……..…………………………..………………………...……..…… 101
Overall limitations………….…………….………………………………...……………..… 102
OVERALL CONCLUSION……..……………………………………………...….…..…… 103
REFERENCES…………..……..……………………………………………...……..…… 104
APPENDIX A. CBCL DSM-V ORIENTED SCALES…………………………....…….. 121
APPENDIX B. UPPS-P POSITIVE AND NEGATIVE URGENCY QUESTIONNAIRE 122

Ra, Dissertation
4

4
List of Figures
Figure 2.1 Conceptual Framework of Study 1…………………………………………………………… 27
Figure 2.2 MATCH EMA PA Item Screenshots………………………………………………………… 31
Figure 2.3 MATCH EMA NA Item Screenshots……………………………………………………… 32
Figure 2.4 Correlation between the mean level of affective states and ADHD………………… 36
Figure 2.5 Correlation between the mean level and variability of affective states ……....…… 37
Figure 2.6 Interaction between the Mean and Variability of Negative Affect Predicting ADHD………. 39
Figure 3.1 Conceptual Framework of Study 2…………………………………………………………… 49
Figure 3.2 Study accrual flow chart …………………………………...………………………………… 52
Figure 3.3 14 items included in digital media activities………………………………………………… 54
Figure 3.4 ADHD prevalence at each follow-ups………………………….…………………………… 61
Figure 3.5 ADHD mean prevalence across all follow-ups……………………………………………… 61
Figure 4.1 Conceptual Framework of Study 3, Aim 5…………………………………………………… 76
Figure 4.2 Conceptual Framework of Study 3, Aim 6…………………………………………………… 77
Figure 4.3 Study Accrual Flow Chart…………………………………………………………….……… 79

Ra, Dissertation
5

5
List of Tables
Table 2.1 Correlations between ADHD symptom criteria and sociodemographic characteristics...…….. 32

Table 2.2 Descriptive Statistics of Sample Characteristics………………………………………………. 35

Table 2.3 Results from the main effect models of affective states and ADHD symptom status.. 36
Table 2.4 Results from the models of moderating effects of variability of affective states in the00
association between affective states and ADHD symptom status………………………...…………... 38
Table 3.1 Descriptive Statistics of Sample Characteristics………………………………………………. 59
Table 3.2 Association of baseline digital media use with ADHD status across 6- 12- 18- and 24-month00
follow-ups among ADHD-negative adolescents………………………………………………………… 63
Table 3.3 Moderating effects of positive urgency in the association of baseline digital media use with00
ADHD status across 6- 12- 18- and 24-month follow-ups among ADHD-negative adolescents……… 65
Table 3.4 Moderating effects of negative urgency in the association of baseline digital media use with00
ADHD status across 6- 12- 18- and 24-month follow-ups among ADHD-negative adolescents……… 67
Table 4.1 Sociodemographic characteristics of samples...………………………………………………. 88
Table 4.2 Results from the models of moderating effects of MVPA in the association between affective00
states and ADHD symptom status……………………………………………………………………... 89
Table 4.3 Sample characteristics at baseline by baseline ADHD status…………………………………. 90
Table 4.4 Moderating effects of MVPA in the association of baseline digital media use with ADHD00
status across follow-ups among ADHD-negative adolescents…….....................………………… 92

Ra, Dissertation
6

6
List of Abbreviations
ADHD Attention deficit/hyperactivity disorder
PA Positive Affect (Affective States)
NA Negative Affect (Affective States)
PU Positive Urgency (Affect Trait)
NU Negative Urgency (Affect Trait)
EMA Ecological Momentary Assessment
DMU Digital Media Use
MVPA Moderate-to-Vigorous Physical Activity
MATCH Study Mother’s and their Children’s Health (MATCH) Study
H&H Study the Happiness & Health (H&H) Study

Ra, Dissertation
7

7
ABSTRACT
Background: The prevalence of ADHD diagnosed in the United States has been constantly
growing among youth in the last two decades. ADHD is one of the most common and high-
impact neurodevelopmental disorders during childhood and adolescence. Thus, it is crucial that
early identification and intervention of ADHD among at-risk populations may alter the trajectory
and long-term consequences this disorder. This study examined the relationships among affect,
digital media use, physical activity, and ADHD symptoms, using real-time data capture
techniques, and longitudinal study design. The three studies addressed (1) the association
between affective states (the mean level and variability of positive and negative affect; PA and
NA) and ADHD among children; (2) the association between digital media use and ADHD in
adolescents, and the moderating effects of affective traits (positive and negative urgency; PU and
NU) between the two; and (3) the moderating role of physical activity in the association between
the mean level of affective states and ADHD and in the association between digital media use
and ADHD.
Methods: Two different datasets including real-time data capture techniques and longitudinal
study design were used to address the research questions. Study 1 adopted a two-stage data
analysis approach using the mean level of affective states (PA and NA) as a predictor, the
variability of PA and NA as a moderator, and ADHD symptom status as an outcome. Study 2
used generalized estimating equations, utilizing repeated measures logistic regressions to assess
the prospective association of digital media use (a predictor), urgency (PU and NU; a moderator),
and ADHD (an outcome). Study 3 used both models to examine how physical activity
(Moderate-to-Vigorous Physical Activity, MVPA) moderates the association between affect and
ADHD and the association between digital media use and ADHD.
Results: The association between the children’s mean level of PA and ADHD symptom status
was negative whilst the association between the children’s mean level of NA and ADHD
symptom status was positive. However, neither association was statistically significant. Whereas
the individual variability of PA did not moderate the association between the children’s mean
level of PA and ADHD symptom status, the individual variability of NA did moderate the
association between the children’s mean level of NA and ADHD symptom status. The frequent
use of modern digital media was associated with increased risk of ADHD in two year follow-ups
among adolescent who were not positive for ADHD at baseline. Moreover, both positive and
negative urgency were positively associated with ADHD symptom status, however, neither
positive nor negative urgency moderated the impact of the frequency of digital media use on
ADHD symptom occurrence among adolescents. The moderating effects of MVPA in the
association between affective states (both PA and NA) and ADHD symptom status and in the
association between digital media use and ADHD symptom status were not significant.
Conclusion: The research conducted in this dissertation contributes to the overall understanding
of the dynamic relationships between affect, digital media use, physical activity, and ADHD in
youth. The findings from these studies have important implications for theory, future research
methods, and ADHD prevention and intervention programs and policies.

Ra, Dissertation
8

8
SPECIFIC AIMS
Three dissertation studies examine the relationship between affect and ADHD symptoms,
using real-time data capture techniques, and longitudinal study design. The three studies address
(1) the association between affective states (the mean level and variability of positive and
negative affect; PA/NA) and ADHD among children; (2) the association between digital media
use and ADHD in adolescents, and the moderating effects of affective traits (positive and
negative urgency; PU/NU) between the two; and (3) the moderating role of physical activity in
the association between the mean level of affective states and ADHD and in the association
between digital media use and ADHD. The specific aims of this dissertation are as follows:
Study 1: Exploring the association of affective states among children and their ADHD
symptom status
Aim 1: To test whether the mean level of affect (PA/NA) is directly associated with children’s
ADHD symptom status.
Hypothesis 1: The mean level of PA is positively associated to children’s ADHD
symptoms.
Hypothesis 2: The mean level of NA is positively associated to children’s ADHD
symptoms.
Aim 2: To test whether the variability of affect (PA/NA) moderates the association of the mean
level of affect among youth and their ADHD symptom status.
Hypothesis 3: The association between the mean level of PA and children’s ADHD
symptom status is moderated by the variability of PA
Ra, Dissertation
9

9
Hypothesis 4: The association between the mean level of NA and children’s symptom
status is moderated by the variability of NA
Study 2: Assessing the association between digital media use and ADHD symptom status
and the moderating effects of affective traits between the two among adolescents.
Aim 3: To test whether digital media use (DMU) predicts the increased risk of ADHD symptom
status.
Hypothesis 5: DMU is positively associated to adolescents’ ADHD symptom status.
Aim 4: To test whether affective traits moderates the association between DMU and ADHD
symptom status.
Hypothesis 6: The association DMU with adolescents’ ADHD symptom status is
moderated by PU.
Hypothesis 7: The association DMU with adolescents’ ADHD symptom status is
moderated by NU.
Study 3: Examining the moderating effects of physical activity in the association between
affective states and ADHD and between digital media use and ADHD.
Aim 5: To test whether physical activity (MVPA) moderates the association of the mean of
affective states among children and their ADHD symptom status.
Hypothesis 8: The association between the mean level of PA and children’s ADHD
symptom status is moderated by children’s MVPA.
Ra, Dissertation
10

10
Hypothesis 9: The association between mean level of NA and children’s ADHD
symptom status is moderated by children’s MVPA.
Aim 6: To test whether MVPA moderates the association between DMU and ADHD.
Hypothesis 10: The association DMU with adolescents’ ADHD symptom status is
moderated by MVPA.

Ra, Dissertation
11

11
CHAPTER 1: BACKGROUND AND SIGNIFICANCE
Attention Deficit/Hyperactivity Disorder (ADHD) is a highly prevalent disorder
diagnosed in youth (Danielson et al., 2018; Perou et al., 2013). ADHD is a neuro-behavioral
challenge characterized by two dimensions (hyperactivity-impulsivity and/or inattention
symptoms) and is associated with affect dysregulation (Barkeley, 1997; Okado & Mueller, 2016).
There are three subtypes of ADHD based on those two dimensions:1) predominantly
hyperactive/impulsive subtype, where adolescents exhibit few or no inattentive symptoms, 2)
predominantly inattentive subtypes, where adolescents exhibit few or no hyperactive/impulsive
symptoms, and 3) combined subtype, where both inattentive and hyperactive symptoms are
present among adolescents (Harpin, 2017; Nigg, 2006). However, these types should only be
seen as descriptive at that point of time since one is likely to have difficulties within different
symptoms at different times in their development (Harpin, 2017).
ADHD is a highly prevalent and high-impact neurodevelopmental disorder in youth.
Even with successful treatment during childhood, the chronic and impairing course of ADHD
throughout the lifespan continues to create a significant burden on the lives of patients and their
families, as well as society. Youth with ADHD are more likely to experience many of
adverseoutcomes compared to those without ADHD (Danielson et al., 2018). These
consequences include lower academic attainment, impaired social functioning, substance use and
gambling, and lead to low income and lower quality of life in adulthood (Danielson et al., 2018;
Fleming et al., 2017; Fletcher, 2014; Groenman, Janssen, & Oosterlaan, 2017; Molina et al.,
2013; Ros & Graziano, 2017). Thus, it is crucial that early identification and earlier interventions
of ADHD among at-risk populations may alter the trajectory and long-term consequences this
disorder (Halperin, Bé dard, & Curchack-Lichtin, 2012).
Ra, Dissertation
12

12
Prevalence of ADHD
There is no global consensus on prevalence estimates for ADHD; however, a systematic
review of prevalence studies estimated the worldwide ADHD prevalence is between 5.29% and
7.1% in children and adolescents (G. Polanczyk, De Lima, Horta, Biederman, & Rohde, 2007; G.
V. Polanczyk, Willcutt, Salum, Kieling, & Rohde, 2014). The prevalence of ADHD in the
United States diagnosed approximately 11% of U.S. youth in 2011, which is approximately two
million more U.S. youth diagnosed with ADHD compared to 2003 (Visser et al., 2014),
suggesting an increasing burden of ADHD among youths in the United States. Thus, efforts to
further understand protective and risk factors associated with ADHD onset are warranted.
Definition of Affect
Affective states, emotion, mood, and feeling are commonly used to describe the
psychological states of an individual. It has long been argued whether either affect or emotion
can be a broader term that encompasses all the other terms (Rahe, Rubin, & Gunderson, 1972;
Schwarz & Clore, 2007). For instance, in psychology and psychiatry research, affective states,
or affect, are used as a broader term whilst some argue that emotion is a more generic term
with respect to its etiology (Lazarus, 1991). In public health and other behavioral science fields,
these terms are often used interchangeably to describe emotional experiences and subjective
feelings (Cyders & Smith, 2007). In the dissertation studies, affect (in general) and emotion
were used synonymously as broader terms to describe emotional experiences and subjective
feelings. Meanwhile, affective states only refer to an individual’s momentary feelings and
experience states.

Ra, Dissertation
13

13
Dimensions of Affective States: Positive Affect and Negative Affect
There are two commonly used approaches for quantifying affective states – 1) the
discrete approach (e.g. fear, anger, pleasure and happiness) and 2) the dimensional approach
(e.g. pleasant-unpleasant and high-low). The differences between these two approaches
mainly stem from the different definitions of emotion from the functionalistic and the
constructive views. First, the discrete approach views affective states as products of
individual appraisal of experiences at the moment of inquiry (Westbrook, 1987). In this
approach, each component of affective states is unique since they correspond to different
reactions to the perceived impact of an event. The dimensional approach is based on the
circumplex model of affect that categorizes affective states into dimensions of valence (i.e.
pleasant or unpleasant) and into states of arousal (e.g. high or low), resulting in a global
value for four distinctively different types of affective states (e.g. ‘feeling excited’ involves
high arousal and high pleasure, and ‘feeling depressed’ involves low arousal and low
pleasure) (Rusell, 1980). In Watson and Tellegen’s research (1985), they first used two
major dimensions derived from analyses of emotions, namely positive and negative affect.
Positive affect refers to feelings such as alertness and activeness, whereas negative affect
refers to unpleasant affective states such as feeling anxious and depressed. In the dissertation
studies, we adopted this approach to refer to dimensions of affective states – positive affect
(PA) and negative affect (NA).

Ra, Dissertation
14

14
Positive and Negative Urgency as Affect Traits
Positive urgency (PU) refers to the tendency to engage in rash actions in response to
positive affective states whilst negative urgency (NU) refers to the tendency to engage in rash
actions in response to affective states (Cyders & Smith, 2008; Marmorstein, 2013; Pedersen et
al., 2016). Impulsivity consists of five components – lack of planning, lack of perseverance,
sensation seeking, negative urgency and positive urgency (Pedersen et al., 2016). The two
personality or affect traits of PU and NU are widely used in research. In the dissertation studies,
urgency (PU and NU) refers to an individual’s affect traits.
Underlying mechanisms between Affective States and Positive and Negative Urgency
Emotional processing has been recognized to prepare the body for action (Davis-Becker,
Peterson, & Fischer, 2014). The relationship between emotional experiences and actions has
been elucidated through activation of the motor cortex by limbic structures (Davis-Becker et al.,
2014). Davis-Becker et al. (2014) stated, “Affect’s facilitation of action is fundamentally
adaptive – affect lead an individual to focus on one particular set of needs out of all the
possible stimuli to which one could conceivably attend.” (Davis-Becker et al., 2014) One takes
action to meet the need identified by the affect. To the degree that affect facilitate actions to
meet needs, it may be true that more intense needs tend to be associated with the experience of
more extreme affective states (Davis-Becker et al., 2014). That is, the experience of more
extreme emotions is likely to be associated with more pronounced needs; and unusual
behavioral choices. For instance, negative experiences of fear may lead an individual to take
radical steps to alter the current situation such as ‘flight vs. fight’. On the other hand, intense
positive experiences of sexual need may propel an individual to take the risky steps of reaching
out to someone to initiate a romantic relationship (Davis-Becker et al., 2014). Thus, the
Ra, Dissertation
15

15
experience of intense emotions tend to interfere with rational thinking and may lead one to
focus more heavily on one’s immediate situation.
Possible Mechanisms Linking Affect with ADHD
ADHD is often explained as a problem of cognitive control, and recently, it has been
argued that ADHD might be better understood by taking an affect regulation perspective (Martel,
2009). The basic argument of considering ADHD as a problem with the regulation of affect is
that affect is not only affected by behavior but also it regulates behavior. That is, affect is
involved in modifying internal feeling states including motivation (Eisenberg & Spinard, 2009).
This is important because affect regulation is crucial to the achievement of personal goals, and a
difficulty in achieving personal goals in a characteristic of ADHD.
There are a number of brain regions and neurotransmitters that have been implicated in
both affective processes and cognitive control. Negative affect, such as fear and anger, appears
to rely on a neural circuit involving the amygdala, hippocampus, anterior cingulate cortex, and
right prefrontal cortex with particular reliance on serotonin neurotransmission (Martel, 2009).
Positive affect relies on the amygdala, the nucleus accumbens, the anterior cingulate cortex, and
the left prefrontal cortex with particular reliance on dopaminergic neurotransmission (Depue &
Lenzenweger, 2006; Derryberry & Tucker, 2006; Fox, 1994; Rothbart & Posner, 2006; Whittle
et al., 2006). Although negative and positive affect is evident during infancy, control processes
develop later during childhood (Casey, Tottenham, Liston,&Durston, 2005). Control processes
rely heavily on the prefrontal cortex, especially the dorsolateral, orbitofrontal, and anterior
cingulate cortex (Derryberry & Tucker, 2006; Rothbart & Posner, 2006; Whittle et al., 2006).
Further, a variety of neurotransmitters are associated with integrity in control processes,
including acetylcholine, norepinephrine, dopamine, and serotonin (Depue&Lenzenweger, 2006;
Ra, Dissertation
16

16
Rothbart & Posner, 2006; Whittle et al., 2006; Martel, 2009). Recent work suggests that the
prefrontal cortex, particularly the anterior cingulate (with dopaminergic and serotoninergic
neurotransmission involvement), likely functions to integrate affect and cognitive control
processes (Holroyd & Coles, 2002). In this way, affect and cognition appear to interact and may
work together to regulate goal-related behavior, such as evaluation of information and execution
of action (Bell & Wolfe, 2004).
Affect regulation is critical for forming basic regulatory abilities (Martel, 2009). Deficits
in affect regulation during the first few years of life may put the child on a trajectory for further
dysregulation, exacerbation of ADHD symptoms, and the development of comorbid conditions.
ADHD follows a developmental trajectory characterized by difficulties with impulse control and
attention that arise early in life (Olson, 1996). Owing to its early development, initial evidence
have suggested that youth with ADHD are often characterized by extreme affect like positive
affect, negative affect and affect volatility (Martel & Nigg, 2006; Nigg et al., 2002a, 2002b).
Possible Mechanisms Linking Digital Media Use with ADHD
Digital media devices are increasingly ubiquitous worldwide (Romer & Moreno, 2017).

In the U.S. for example, 92% of adolescents reported going online daily in 2015— including 24%
who reported being online “almost constantly.”(Lenhart et al., 2015) Smartphones and other
mobile devices provide instant and constant access to numerous modern digital media activities
that are highly popular amongst youth, such as social networking, streaming movies or music,
texting, video chatting, and mobile videogames (Lenhart et al., 2015). Increasing digital media
use has raised concerns amongst health professionals regarding the neurobehavioral
consequences of excessive digital media exposure in youth (George & Odgers, 2015).
Ra, Dissertation
17

17
As aforementioned, attention deficit/hyperactivity disorder (ADHD) is a psychiatric
condition involving persistent difficulty sustaining attention, hyperactivity, and impulsivity that
is associated with adverse educational, social, legal, and health outcomes (Nigg, 2013). The
prevalence of ADHD is approximately 5% to 7% of the global population of youth as of 2017
(Nyarko et al., 2017) and is still increasing in some regions (Atladottir et al., 2015). While
ADHD has been historically considered a childhood disorder, it has also been found that an
considerable portion of ADHD cases onset during adolescence and persist throughout adulthood
(Moffitt et al., 2015), which means that the course of ADHD may be associated with
environmental effects that emerge in adolescence (Larsson, Larsson, & Lichtenstein, 2004).
The relationship between digital media use and risk of ADHD in adolescence can be
elucidated by: (1) the scan-and-shift hypothesis; and (2) the fast-pace arousal-habituation
hypothesis (Nikken & de Haan, 2015). The scan-and-shift hypothesis refers to promoting an
attentional style of rapid scanning and shifting that inhibits sustained attention such as reading a
book (Nikken & de Haan, 2015). The fast-pace arousal-habituation hypothesis denotes the
increased arousal by fast pace of digital media use, which inhibits ability to sustain focus
(Nikken & de Haan, 2015). This could also habituate youth to immediate feedback and
stimulation, and in turn stunt the development of impulse control (Nikken & de Haan, 2015).
While use of older digital media platforms, such as television viewing and console video
game playing, has been linked with ADHD in youth (Nikken & de Haan, 2015), it is unknown
whether the integrated use of modern forms of digital media including social networking or
texting is associated with ADHD in adolescence.
Ra, Dissertation
18

18
Adolescence is an important time for social cognitive development. Subcortical regions
functionally associated with emotion processing and reward undergo considerable changes and
reorganization during puberty (Brenhouse & Andersen, 2011). The dopaminergic system and
related regions in the striatum are implicated in potential mechanisms underlying two common
features of adolescence: escalation in risk-taking behaviors and increased desire to spend time
with and earn the approval of peers (Steinberg, 2008). Modern digital media use may promote
those features of adolescence. For instance, viewing photos with high attention on the social
media use is associated with greater activity in neural regions implicated in reward processing
and attention, whereas viewing photos of risky behaviors (e.g. smoking, drinking) decreased
activation in the cognitive-control network (George & Odgers, 2015). Likewise, there has been
growing public health concern that increasing digital media usage may adversely impact
population neurobehavioral health such as ADHD.
Underlying Mechanisms between ADHD and Physical Activity
Previous studies have found that physical activity has positive impacts on many of the
same neuro-biological factors associated to ADHD (Gapin, Labban, & Etnier, 2011). A few
studies have found that physical activity increases cerebral blood flow (Swain et al., 2003), and
the availability of dopamine and norepinephrine in synaptic clefts of the central nervous system
(Fulk et al., 2004). Physical activity was also found to lead changes in cerebral structure that are
known to be important for cognitive performance (Stanley J Colcombe et al., 2006; Kramer,
Erickson, & Colcombe, 2006; Van Praag, 2008). Moreover, greater brain activity within regions
of cortex were found to be associated with behavioral conflicts and attentional control process
(Stan J Colcombe, Kramer, McAuley, Erickson, & Scalf, 2004). Therefore, there is evidence that
Ra, Dissertation
19

19
physical activity benefits cognitive functions, which interact with affect, in turn, may ameliorate
the symptoms of ADHD (Gapin et al., 2011).
Current Research Gaps
Previous studies on affective states and ADHD have several methodological limitations
(e.g. using self-reported affect using survey questions). Of those, no research has been conducted
on the potential moderating effect of variability in affective states in the relationship between the
mean level of affective states and ADHD symptom status.
While Modern digital media platforms are considered more neuropathogenic than
traditional digital media such as television viewing due to their high speed, level of stimulation
and ubiquitous use throughout the day. (George & Odgers, 2015). However, the evidence to date
was insufficient to address this question because of inadequate assessment of digital media use
and poor study designs to test causal inferences (Pediatrics, 2017). In addition, almost no studies
have explored a role of urgency (PU or NU) in the relationship between the digital media use and
ADHD.
Meanwhile, while there is a general consensus regarding the benefits of physical activity
to mental health such as ADHD, whether its moderating role in the course of ADHD is not clear.
Especially, the moderating effects of physical activity in the association between affective states
and ADHD, and between digital media use and ADHD are not well explored.

Ra, Dissertation
20

20
Overview of dissertation studies
The dissertation studies seek to examine the dynamic relationships between affects,
digital media use, physical activity and ADHD in children and adolescents, using real-time data
capture techniques (data from the Mother’s and their Children’s Health (MATCH) Study) and
longitudinal study design (data from the Happiness & Health (H&H) Study). These analyses use
two different datasets to address different research questions in two populations at risk for
ADHD – children (8-12 year-old) and adolescents (15-18 year-old).
Study 1 used the Mother’s and their Children’s Health (MATCH) Study, a longitudinal
study of mother-child dyads who were given six semi-annual assessments over the course of
three years. The MATCH Study uses real-time data capture methodology to examine the within-
day and long-term effects of affective states, physical activity, and mental health risk including
ADHD. Using a novel two-stage data analysis approach, Study 1 investigated the effects of
Ecological Momentary Assessment (EMA) measured affective states (PA and NA) on ADHD
symptom status. We first examined the relationship between the individual mean level of PA and
NA and ADHD symptom status, then further explored the moderating effects of the individual
variability of PA and NA in the association between the two using data collected from wave 1 of
the MATCH study.
Study 2 analyzed data from the Happiness & Health (H&H) Study, a longitudinal survey
of health behaviors and mental health amongst students enrolled in ten public high schools in
Los Angeles, CA, USA, with six semi-annual assessments given over four years, (Leventhal et
al., 2015). In Study 2, we examined 1) the association between baseline use of modern digital
media (wave 3) and subsequent likelihood of meeting symptom criteria for ADHD (wave 4-8),
Ra, Dissertation
21

21
and 2) the moderating effects of positive urgency (PU) and negative urgency (NU) between the
association of the two across a 30-month follow-up period in adolescents who did not meet
symptom criteria for ADHD at baseline.
Study 3 used both MATCH and H&H data to explore the moderating role of physical
activity: 1) on the association between mean level of affective states and ADHD identified in
Study 1; and, 2) on the association between digital media use and ADHD identified in Study 2.
These dissertation studies on the prospective association between affects, digital media use,
physical activity, and ADHD in youth will extend the literature on ADHD and provide critical
information about pathways of ADHD in children and adolescents.

Ra, Dissertation
22

22
CHAPTER 2: EXPLORING THE ASSOCIATION OF AFFECTIVE STATES AMONG
CHILDREN AND THEIR ADHD SYMPTOM STATUS
INTRODUCTION
Attention deficit/hyperactivity disorder (ADHD) is one of the most common disorders
diagnosed in childhood (Danielson et al., 2018; Perou et al., 2013). The prevalence of ADHD in
the United States youth was affected approximately 11% in 2011, which is approximately two
million more, compared to the 7.8% of children with ADHD in 2003 (Visser et al., 2014).
ADHD is a high –impact neurodevelopmental disorder of childhood (Barkley, 1997). Children
with ADHD are more likely to experience a variety of negative outcomes compared to those
without ADHD (Danielson et al., 2018). Effects include lower academic attainment, impaired
social functioning, substance use and gambling, which may lead to low income and low quality
of life in adulthood (Danielson et al., 2018; Fleming et al., 2017; Fletcher, 2014; Groenman,
Janssen, & Oosterlaan, 2017; Molina et al., 2013; Ros & Graziano, 2017). Children with ADHD
experience serious functional deficits across domains, which may persist during adolescence and
throughout adulthood (Moffitt et al., 2015). Thus, addressing risk factors earlier on is important
to prevent severe symptoms of ADHD and to minimize its life-long impact.
Despite dramatic progress made in ADHD research in the past decade, the critical role of
affect as it relates to the course of ADHD has been lacking (Brown & Brown, 2014). Affect are
critically important motivators of human thoughts and actions, therefore understanding the
essential role of emotion should be integrated into ADHD research (LeDoux, 1996).

Ra, Dissertation
23

23
The Mean Level and Variability Of Affective States (Positive Affect And Negative Affect)
In Mental Health Research
Prior to addressing the importance of affect in understanding ADHD, it is essential to
define what affect is. In previous studies, affect refers to either 1) trait affect (trait positive affect
and trait negative affect) or 2) affective states (positive affective states and negative affective
states). Trait negative affect denotes individuals with greater propensity to experience more
intense and frequent negative emotions while trait positive affect refers to the tendency to
experience positive affect (Hamilton et al., 2017). On the other hand, affective states (both
positive and negative) refer to the immediate positive or negative emotional states that an
individual experiences at a momentary time point (Genevieve Fridlund Dunton et al., 2014).
Positive affect and negative affect in the dissertation studies refer to positive affective states (PA)
and negative affective states (NA) respectively. The variability of affect refers to fluctuations of
affective states (either PA or NA) within-individuals across time in response to pleasant events,
stressors, or other stimuli at a given time (Maher, 2018). The mean level of affect states the
average of affective states (either PA or NA) within-individuals across time in response to
pleasant events, stressors, or other stimuli at a given time (Maher, 2018). While trait affect is
personal-level characteristics that individuals have long-term tendencies toward experiencing
affect in a variety of situation, the variability of affective states is within-person momentary
mood changes (Brondolo et al., 2008) . The mean level and variability of affective states may
differ between people, and thus may refer to a source of between-person variation.

Ra, Dissertation
24

24
Lack of Consensus of Whether the Relation Of Affect Mean and Variability with ADHD Is
Positive or Negative
Recently, some attention has been paid to the association between ADHD and affect, and
of a few studies that have been conducted, the findings were mixed. As aforementioned, affect
can be classified into two dimensions – positive affect (PA) and negative affect (NA). Only a
few studies have examined the relationship between children’s mean level of PA/NA and ADHD.
According to Loney and his colleagues (2006), the hyperactivity/impulsivity symptom subscale
of ADHD was positively related to NA but not PA. Baldwin and Dadds (2008) found that
ADHD symptoms were positively associated with parent- and child-reported NA. In addition,
one study reported that negative affect is more likely to persist in children with ADHD (Whalen
et al., 2009). On the other hand, PA has been considered to trigger the hyperactivity and
impulsivity dimension of ADHD (Herzhoff et al., 2013). Moreover, high behavioral approach
motivation tendencies inherent in PA have been found to predict hyperactive-impulsive
symptoms in young adults with ADHD (Mitchell, 2010; Okado et al., 2013). A recent clinic-
based study (Okado et al., 2013) reported that children with ADHD reported higher levels of PA
than other referred children in a clinic-based study. Together, these findings suggest that both
high levels of NA and PA may be associated with ADHD symptomatology in youth.
Rather than separately examining the relationship between the level of PA/NA and
ADHD, many studies have focused on the aggregated affect variability (also referred to as
variability in emotion/mood, emotional instability, emotion dysregulation, affect volatility, and
mood liability) and its association with ADHD (Kearnes & Ruebel, 2011). One recent study
among 102 children (56 with ADHD and 46 without ADHD) reported that higher affect
variability (of combined PA and NA) is associated with behavioral problems and the association
Ra, Dissertation
25

25
is stronger in those with ADHD (Paul J. Rosen, Walerius, Fogleman, & Factor, 2015). Other
studies have supported the notion that both more PA and NA variabilities are related to
problematic behaviors (e.g. externalizing behavior) and mental health diagnoses including
ADHD (Factor, Reyes, & Rosen, 2014; Paul J Rosen, Epstein, & Van Orden, 2013; Paul J.
Rosen et al., 2015; Whalen et al., 2009). Evidence from the studies described in this section
provides preliminary evidence that both mean PA and NA, as well as overall affective variability
may be uniquely associated with ADHD in children.
Variability of Affect As a Possible Moderator of the Association Between the Mean Level
Of Affect (PA/NA) and ADHD
As aforementioned, previous studies have examined the association between the mean
level of affective states and ADHD; and the association between the affect variability and ADHD
separately. To date, no study to date has explored the moderating effects of affect variability on
the association between the mean level of affect (i.e. PA/NA) and ADHD. Those with high mean
and high variability of positive affect may be more likely to have higher risk of mental illness
such as ADHD, which is characterized by recurrent states of high positive affect (Gestsdottir &
Lerner, 2008; Gruber, Johnson, Oveis, & Keltner, 2008). A recent study found that even if there
was a negative relationship between mean of PA and depressive symptoms, the effects of mean
PA was attenuated when the variability of PA is higher than the average (Maher et al., 2018). In
support of this view, there might be a differential association between the mean level of PA/NA
and ADHD based on an individual’s magnitude of affect variability (i.e. moderating effect).

Ra, Dissertation
26

26
Current Research Gaps
Although examining the association between affect and ADHD is not completely new,
previous studies have a number of limitations. First of all, little is known regarding the
dimensions of affective states and its impact on ADHD in youth. Thus, a specific role of affect
variability in relation to ADHD should be examined separately. On the other hand, previous
studies examining on the relationship between ADHD and the mean level of affect often did not
differentiate between the two separate dimensions of affect (i.e. PA/NA). Furthermore, none of
the existing studies have explored the potential the moderating effects of affect variability
between the mean level of affective states and ADHD. In addition, most studies used
retrospective designs, addressing ADHD by predicting the mean level of affect or affect
variability using case-control study (i.e., ADHD vs. no ADHD). Most previous research assessed
affect by survey questions such as PANA-C, thus was not able to detect the momentary changes
or variations. Lastly, no studies have explored the interaction between the mean level of affect
and the variability of affect in course of ADHD, which may yield new insights into the role of
affect and its variability in ADHD, which may yield new insights into the role of affect and its
variability in ADHD.
Specific Aims and Hypotheses
This study aimed to investigate the effects of EMA measured affective states (PA and NA)
on ADHD symptom status in great details by adopting a novel statistical approach. We first
separately examined the relationship between the children’s mean level of PA (and NA) and
ADHD symptom status. Next, we explored the moderating effects of the individual variability of
PA (and NA) in the association between the mean affect and ADHD symptom status using the
Ra, Dissertation
27

27
wave 1 of MATCH study. To test these, the prompt-level and the person-level data were merged.
The basic conceptual model is displayed in Figure 2.1.

Figure 2.1 Conceptual framework of Study 1
Aim 1: To test whether the mean level of affect (PA/NA) is directly associated with children’s
ADHD symptom status.
Hypothesis 1: The mean level of PA is positively associated with children’s ADHD
symptoms.
Hypothesis 2: The mean level of NA is positively associated with children’s ADHD
symptoms.
Aim 2: To test whether the variability of affect (PA/NA) moderates the association between
mean level of affect and ADHD symptom status in children.
Hypothesis 3: The association between the mean level of PA and children’s Symptom
status is moderated by variability of PA (i.e. the mean level of PA and ADHD status is
Ra, Dissertation
28

28
positively associated at higher variability of PA whereas the mean level of PA and
ADHD status is negatively associated at lower variability of PA).
Hypothesis 4: The association between mean level of NA and children’s Symptom status
is moderated by variability of NA (i.e. the mean level of NA and ADHD status is
positively associated at higher variability of NA whereas the mean level of NA and
ADHD status is negatively associated at lower variability of NA).
METHODS
Data, Participants and Procedures
This research used data from the Mothers’ and Their Children’s Health (MATCH) Study
(PI: Dr. Genevieve Dunton). The MATCH Study is a prospective study with a total of 202
mothers and their 8 to 12 year-old children (baseline), measured at 6 semi-annual assessments
(e.g., waves) from the fall of 2014 to the spring of 2018. The inclusion criteria for MATCH
study were: 1) currently in 3rd – 6th grade (child), 2) reside together at least 50% of time (mother
and child), and 3) ability to speak and read in English or Spanish (mother and child). Study
exclusion criteria included: 1) use of medication for thyroid or psychological conditions (mother),
2) a health condition limiting physical ability (mother or child), 3) enrolled in a special education
program (child), 4) currently using oral or inhalant corticosteroids for asthma (mother or child), 5)
pregnancy (mother), 6) underweight (BMI < 5th% for age and sex) (child), and 7) working more
than two evenings (between 5-9 pm) during the week or more than one 8-hour weekend shift
(mother). Each wave consists of accelerometer monitoring; ecological momentary assessment
(EMA); and paper-and-pencil surveys in mothers and children. Accelerometers are wearable
devices using sensors to detect body movements in order to estimate the intensity of activity over
Ra, Dissertation
29

29
time (Chen & DAVID R BASSETT, 2005). EMA is an innovative real-time data collection
methodology that uses electronic devices (e.g., smartphone apps) (Dunton et al., 2015) to
repeatedly assess participants during daily life; EMA methods minimize participants’ recall bias
and improve external and ecological validity. The MATCH Study is the first of its kind to use
EMA to examine the within-day (3 random prompts a day during weekdays and 7 random
prompts during weekends for 1 week) and long-term (6 waves over 3 years) effects of
momentary affect on physical activity, and mental health among mothers and children.
Sociodemographic characteristics were assessed through paper-and-pencil surveys.
Measures
ADHD Symptom Status
The outcome, ADHD, was assessed through mothers’ reports on the Child Behavior
Checklist (CBCL) in wave 1. The CBCL is a paper-pencil instrument with 142 questions. It is
the most widely-used standardized measure for evaluating maladaptive behavioral and emotional
problems in youth, and evidence for content, construct, and criterion-related validity is well
documented (Achenbach & Edelbrock, 1991). It also serves as a screening tool to identify
ADHD cases in a pediatric primary care referred population (Biederman, Monuteaux, Kendrick,
Klein, & Faraone, 2005) since it is efficient in minimizing the amount of time required by the
primary care physician, and is also cost effective in contrast to the highly intensive works of
structured diagnostic interview methods by a physician (Biederman et al., 2005). Seven items
(fails to finish, concentrate, sit still, impulsive, inattentive, talk much, and loud; see Appendix I
for more details) were used to compute the DSM-V ADHD criteria (below borderline clinical
range, within borderline clinical range and above clinical range) based on gender (boy vs. girl)
Ra, Dissertation
30

30
and age (6-11 vs. 12-18). For the current analysis, ADHD criteria category was dichotomized
into below borderline clinical range (0) vs. within or above borderline clinical range (1) to refer
to the current ADHD symptom status. Detailed information on use, scoring is accessible at:
http://www.aseba.org/.
EMA Affective States
The main predictors, children’s affective states (PA and NA), were obtained from EMA
data from wave 1. Five affective state items were adopted from the Positive and Negative Affect
Schedule for Children (PANAS-C; Ebesutani et al., 2012; Laurent et al., 1999). To assess PA,
two questions were asked, “How 1) HAPPY, 2) JOYFUL were you feeling right before the
phone went off?” (Figure 1.2). NA was assessed by 3 questions, “How 1) MAD, 2) SAD, 3)
STRESSED were you feeling right before the phone went off?” (Figure 1.3). Responses were
“1=Not at all”, “2=A little”, “3=Quite a bit”, “4=Extremely”. Across all available EMA reports
for each participant, scores for PA were averaged to create a person-level mean of PA
(Chronbach’s alpha=.75); and scores for NA were averaged to create a person-level mean of NA
(Chronbach’s alpha=.79).
Ra, Dissertation
31

31

Figure 2.2. MATCH EMA PA Item Screenshots

Figure 2.3. MATCH EMA NA Item Screenshots

Ra, Dissertation
32

32
Covariates
Person-level covariates (i.e. ethnicity, annual household income, single parent status, and
mother’s educational attainment) were screened to include in the second stage model. It was
based on significant bivariate correlations with the outcome (p <.05). As a result, adjusted
second stage models controlled for the single parents household status (single-parent vs. dual-
parent). The outcome, the ADHD symptom criteria, was already adjusted for age and gender,
thus were not included in the second stage models to avoid overall-controlling.
Correlations among the ADHD symptom criteria, the mean levels of PA and NA, and
relevant covariates are displayed in Table 2.1.

Table 2.1. Correlations between ADHD symptom criteria and sociodemographic
characteristics
Variables 1 2 3 4 5 6
1. ADHD symptom criteria
a
1.00
2. Mother attended college -.09 1.00
3. Mother work fulltime -.11 .02 1.00
4. Single parent household .24** -.10 .09 1.00
5. Age .005 .00 -.04 .03 1.00
6. Ethnicity -.04 -.33** .003 .07 .10 1.00
7.Gender .05 -.07 -.03 .00 -.11 -.05
Note. Data displayed in this table are based on 180 to 202 subjects, depending on the variable, due to missing data.
ADHD symptom criteria, attended college, work fulltime, single parent household, ethnicity, and gender are
dichotomized variables with having positive ADHD symptoms, attending college, working fulltime, being in a
single-parent household type, being Hispanic, and being a boy.
a
ADHD symptom criteria has been adjusted for age
and gender based on the CBCL scoring rules.
** p <.01

Ra, Dissertation
33

33
Statistical Analysis
Prior to analysis, the prompt-level dataset providing PA and NA at the prompt-level and
the person-level dataset, which include the person-level ADHD outcome and covariates were
merged. Prompt-level NA and PA variables were created by averaging the two PA (Happy and
Joyful) and the three NA (Mad, Sad and Stressed) items. This study employed a two-stage data
analysis approach (MIXREGLS-MIXOR models (Hedeker & Gibbons, 1996, 2006) in which 1)
the first stage uses the prompt-level ratings to calculate the person-level average and the
variability of PA and NA, and 2) the second stage integrated the effects from the first stage
model into a single-level logistic regression using person-level variability in affect to predict
odds of ADHD (see Hedeker & Nordgren, 2013 for more details on this modeling approach).
This novel approach can take into account covariates and varying numbers of observations when
calculating the person-level variability, and then use that person-level variability to, in turn,
predict a person-level outcome, which existing techniques cannot do. The first stage was tested
as an empty model with no covariates, and the second stage controlled for the single parents
household status (single-parent vs. dual-parent). Four separate models were tested –two main
effect models – each looking at the effect of either PA or NA on ADHD, and two interaction
models, looking at the interaction effect of affective variability on the relationship between mean
of PA or NA and ADHD.
Main effect model:
Level 1: PA (or NA)_ij = b0_i+ + e_ij
Level 2: ADHDi = ß 0+ ß 1*Single parent household+ ß
2
*Mean
i
+ ß
3
*Variability
i

Ra, Dissertation
34

34
Interaction model:
Level 1: PA (or NA)_ij = b0_i+ + e_ij
Level 2: ADHDi = ß 0+ ß 1*Single parent household+ ß
2
*Mean
i
+ ß
3
*Variability
i

+ ß
4
* Mean
i
_Variability
i

• i: Person-level, j: prompt-level
RESULTS
Descriptive Statistics
This study used Wave 1 MATCH dataset. Table 2.2 shows baseline sociodemographic
characteristics of the MATCH study (N=202). Among all 202 children, 34 participants (16%)
were excluded due to incomplete data. After excluding 34 participants, 168 participants (84%)
were included in the analytic sample (Table 2.2).

Ra, Dissertation
35

35
Table 2.2. Descriptive Statistics of Sample Characteristics

Baseline Sample
a
Analytic Sample
b
Variable N (%) N (%)
Child Sex
Male 99 (49.0) 90 (53.6)
Female 103 (51.0) 78 (46.4)
Child Age, M (SD) 9.61 (0.91) 9.63 (0.97)
Child Ethnicity
Hispanic 109 (54.0) 92 (54.8)
Non-Hispanic 93 (46.0) 76 (45.2)
Annual Household Income
Less than $35,000 55 (27.2) 44 (26.2)
$35,001-$75,000 59 (29.2) 50 (29.8)
$75,001-$105,000 39 (19.3) 35 (20.8)
$105,001 and above 48 (23.8) 39 (23.2)
Positive Affect
c

Mean (SD) 3.02 (0.68) 3.04 (0.67)
Negative Affect
d

Mean (SD) 1.27 (0.27) 1.26 (0.26)
ADHD symptom status
Below borderline
clinical range
175 (94.5) 158 (94)
Within or above
borderline clinical range
11 (5.5) 10 (6.0)
Total N
202 168
Note.
a
The number of observations does not add up to total due to missing data on that variable.
b
Final analytic sample using listwise deletion of main variables.
c
Score range: 1 (Not at all) to 4
(Extremely) of positive affect (the average of feeling happy and joyful) when prompted.
d
Score
range: 1 (Not at all) to 4 (Extremely) of negative affect (the average of feeling mad, sad and
stressed) when prompted. Abbreviations: SD = Standard deviation; N = Sample size.

There were slightly more girls than boys (53.6% vs. 46.4%) in the analytic sample, and
the mean age of participants was 9.63 years old. Over a half of the sample was Hispanic (54.8%).
The mean of positive affect was 3.04 (3=quite a bit) out of 4, and the mean of negative affect
was 1.26 (1=a little) out of 4. 10 participants (6%) were classified as ‘within or above borderline
clinical range’.
Ra, Dissertation
36

36
Figure 2.4 depicts the relationships between the mean level of affective states and ADHD.
There was a slightly negative relationship between the mean level of positive affect and the
ADHD status, but it was not statistically significant (p=.82) whereas the correlation between the
mean level of negative affect and the ADHD status was positive and was statistically significant
(p=.04).

Note: The mean levels of affective states (PA and NA) were standardized. The ADHD status in the y-axis was a
continuous variable.

Figure 2.4. Correlation between the mean level of affective states and ADHD

Table 2.3. Results from the main effect models of affective states (PA/NA) and ADHD
symptom status
Positive Affect Negative Affect
Model Variable Estimate
Standard
Error
p-
value Estimate
Standard
Error
p-
value
Unadjusted Intercept -2.83 0.34 <.001 -2.95 0.37 <.001
Mean of affect 0.06 0.30 0.82 0.43 0.24 0.08

Adjusted Intercept -3.76 0.59 <.001 -3.82 0.76 <.001
Mean of affect 0.11 0.30 0.71 0.28 0.27 0.3

Single parent
household 2.31 0.74 0.002 2.19 0.81 0.006
Ra, Dissertation
37

37
Based on the results from the main effect model (Table 2.3), the mean level of negative
affect was marginally positively associated with the ADHD symptom status (βMean =0.43, p=.08)
whereas the mean level of positive affect did not significantly predict ADHD symptom status
(p=.82). After adjusting for the single parent household, the association between the mean level
of negative affect and the ADHD symptom status became insignificant (βMean =0.28, p=.3). The
single parent household was significantly associated with the ADHD symptom status in both PA
and NA models (βMean =2.31, p=.002; βMean =2.19, p=.006 respectively).
Figure 2.5. Correlation between the mean level and variability of affective states

Figure 2.5 depicts the relationships between the mean level and the variability of
affective states. There was a negative correlation between the mean level and the variability of
positive affect, which means that participants whose mean level of positive affect is higher than
the average are less likely to have higher variability of positive affect. Interestingly, there was a
positive correlation between the mean level and the variability of negative affect. That is,
participants whose mean level of negative affect is higher than the average are more likely to
have higher variability of negative affect.
Ra, Dissertation
38

38
Table 2.4. Results from the models of moderating effects of variability of affective states in
the association between affective states and ADHD symptom status
Positive Affect Negative Affect
Model Variable Estimate
Standard
Error
p-
value Estimate
Standard
Error
p-
value
Unadjusted Intercept -2.84 0.45 <.001 -3.63 0.61 <.001
Mean of affect 0.01 0.39 0.98 -0.54 0.89 0.55
Variability of affect -0.13 0.43 0.77 0.19 0.73 0.79

Mean x Variability
Interaction -0.32 0.35 0.37 0.85 0.38 0.03

Adjusted Intercept -3.81 0.66 <.001 -4.53 1.04 <.001
Mean of affect 0.04 0.40 0.92 -0.79 1.04 0.45
Variability of affect -0.13 0.48 0.79 0.41 0.89 0.64

Single parent
household 2.25 0.76 0.003 2.24 0.97 0.02

Mean x Variability
Interaction -0.27 0.42 0.52 0.85 0.41 0.04

For the association between positive affect and ADHD symptom status, there was no
moderating effects of variability of positive affect in both unadjusted and adjusted models (p=.37,
p=.52 respectively). However, the variability of negative affect significantly moderated the
association between negative affect and ADHD symptom status (βMean =0.85, p=.03) in the
unadjusted model. The moderating effects of variability of negative affect remained significant
after adjusting for the single parent household (βMean =0.85, p=.04).

Ra, Dissertation
39

39

Figure 2.6. Interaction between the Mean and Variability of Negative Affect Predicting
ADHD

Based on the results from the adjusted model, the main effects of mean level and
variability in NA were not directly associated with ADHD symptom criteria. However, the
association between the mean level of NA and ADHD varied as a function of variability of NA
(βMean =0.85, p=.04). More specifically, there was a positive relationship between the mean level
of NA and ADHD for individuals with higher variability, meaning that children with higher
mean level of NA are more likely to experience ADHD symptom status when their variability is
higher than the average.

-8
-7
-6
-5
-4
-3
-2
-1
0
Low Mean Average Mean High Mean
Average variability
High variability
Low variability
Ra, Dissertation
40

40
DISCUSSION
This study examined 1) the relationship between the children’s mean level of PA (and
NA) and ADHD symptom status and 2) the moderating effects of the individual variability of PA
(and NA) in the association between the mean affect and ADHD symptom status. The
association between the children’s mean level of PA and ADHD symptom status was negative
whilst the association between the children’s mean level of NA and ADHD symptom status was
positive. However, neither association was statistically significant. Whereas the individual
variability of PA did not moderate the association between the children’s mean level of PA and
ADHD symptom status, the individual variability of NA did moderate the association between
the children’s mean level of NA and ADHD symptom status. That is, children with higher mean
level of NA are more likely to have ADHD symptom status when their variability of NA is also
higher than the average.
Results from this study partially support previous research suggesting a positive
association between NA and ADHD. A few studies suggested that ADHD is positively related to
NA but not PA. Loney and his colleagues (2006) found that the hyperactivity/impulsivity
symptom subscale of ADHD was positively related to NA but not PA. Another study reported
that ADHD symptoms were positively associated with both parent- and child-reported NA
(Baldwin & Dadds, 2008). In addition, one study reported that negative affect is more likely to
persist in children with ADHD (Whalen et al., 2009). The results demonstrated a positive
relationship between NA and ADHD, but it was not significant, which may be due to the small
sample size. Thus, future studies should examine the same models using a larger number of
sample (e.g. more number of prompts and/or more participants) to confirm the association and
generalizability.
Ra, Dissertation
41

41
Furthermore, this study provides the first known evidence that the variability of NA
moderates the association between the mean level of NA and ADHD symptom status. Children
whose NA is higher than the average are more likely to have ADHD symptom status only when
their variability is higher than the average. Previous studies examined the association of ADHD
with either the mean level of affective states or the variability in affective states (Al‐Yagon, 2009;
Anestis & Joiner, 2011; Bunford, Evans, & Wymbs, 2015; Factor et al., 2014). The role of
variability in momentary ratings of NA on aspects of mental has been unclear in previous
research (Maher et al., 2018). This study helps to further understand the dynamic association
between affective states and ADHD by exploring the role of variability of affective states using
the novel methodology. Moreover, the novel methodology and analytic approach used in this
article can be applied to different populations and research areas that utilize real-time data
capture methodology. These findings may indicate that future study needs to explore the
dynamic nature of affect and its relationship with mental health in both clinical and nonclinical
populations.
Limitations
Even though the present study is strengthened by the use of electronic EMA to assess
current affective states to measure affective states in free-living settings and adopts the novel
statistical approach to handle the integrated two-stage models, there are some potential
limitations. Due to technical and/or other unknown issues, a few children (n=11) did not comply
with EMA prompts at all. Thus, missing data is a potential limitation due to the possibility of
missing not completely at random (i.e. biased) based on children’s characteristics.
Ra, Dissertation
42

42
There is also evidence that coherence of children’s self-report of their emotion is
positively related to their age (Durbin, 2010), which means that interpretations should be made
with caution given younger children’s self-reports of affect have been found to demonstrate
response biases (Rosen et al., 2013; Okado, Mueller, & Nakamura, 2016). In addition, results
from this study may differ from samples containing ADHD diagnosis. To assess the ADHD
symptom status, we used the DSM-Oriented scales from CBCL. Although there is enough evidence
indicating that the DSM-Oriented scales provide relevant information for the assessment of
functional impairment, and they are used as a complementary screening tool in clinical practice,
responses on the rating scale are not sufficient to render a diagnosis of ADHD, but may be
consistent with the disorder, indicating that further clinical evaluation is necessary.
Lastly, since this study used a cross-sectional data, we cannot assess the temporal
ordering of affective states and ADHD symptom status even though our study is based on the
assumption that affective states are detected prior to ADHD symptom status. Thus, there is
potential for directionality effects.
Implications
This is the first known study to examine the potential moderating effect of variability in
affective states in the relationship between the mean level of affective states and ADHD
symptoms status using a novel methodology. Findings from this study will shed light on the
critical role of affective states to understand course of ADHD, which has been lacking in ADHD
research.

Ra, Dissertation
43

43
By analyzing real-time data collected from mobile phone app in a free-living setting, this
study is able to address several methodological limitations of previous studies on this topic.
Compared to most previous studies using self-report of affect using survey questions such as
PANAS, which have a potential for recall bias, intensive longitudinal designs of EMA repeatedly
capture momentary affect across time, whereby providing a rich data source for modeling
variance in person-level mean and variability parameters across time (Hedeker et al, 2012;
Maher, 2018). More importantly, this modeling approach includes a random subject effect for the
within-subject variance specification, which allows the within-subject variance to vary at the
subject level in order to be modeled as a person-level predictor, along with person-level mean
(Hedeker & Nordgren, 2013; Maher, 2018).
As aforementioned, affect is critically important motivators of human thoughts and
actions, thus understanding the essential role of affect should be integrated into ADHD research.
This study first explored the interactive association between the mean level and variability of
affective states in course of ADHD symptom status. This is novel because previous studies have
only addressed the role of either the mean level or variability of affective states on ADHD
(Okado et al., 2016). The results from this study will help to further elucidate the differentiated
association between affective states and ADHD. Thus, this investigation is important to better
understand the dynamic nature of affect and its relationship with ADHD.
Most previous studies on affect and ADHD have tested ADHD diagnosed patients using
clinical samples and focused on the post-treatments (e.g. van Stralen, 2016). However, even
though it is successfully treated in childhood, the chronic and impairing course of ADHD
throughout the lifespan continues to negatively impact the lives of patients, their families, as well
as society. Yet, compelling arguments have been made for the use of earlier interventions which
Ra, Dissertation
44

44
may alter the trajectory of the disorder and thus avoid a variety of the long-term consequences of
ADHD (Halperin et al., 2012). Early intervention is essential in addressing an array of
neurodevelopmental disorders such as ADHD since childhood is a critical period for brain
development.
Future Research
Further research using six waves of MATCH data should be employed to examine if the
moderating effects of variability of negative affect between the mean level of negative affect and
ADHD symptom criteria remain the same.
CONCLUSION
The findings from this study contributes to identify at-risk children prior to the onset of
serious ADHD symptomatology, and provide secondary preventive interventions (e.g. parent
training for children at higher risk of ADHD) to reduce (or eliminate)
the likelihood of severe consequences of ADHD in the future. For example,
understanding how a child’s mean level and variability in negative affect relate to ADHD risk
can aid in the early detection of ADHD symptomatology, leading to earlier intervention and
improved outcomes.

Ra, Dissertation
45

45
CHAPTER 3: ASSESSING THE ASSOCIATION BETWEEN DIGITAL MEDIA USE
AND ADHD SYMPTOM STATUS AND THE MODERATING EFFECTS OF
AFFECTIVE TRAITS BETWEEN THE TWO AMONG ADOLESCENTS
INTRODUCTION
ADHD is a highly prevalent and high-impact neurodevelopmental disorder during
childhood and adolescence (Ra et al., 2018). Even with successful treatment during childhood,
the chronic and impairing course of ADHD throughout the lifespan continues to create a
significant burden on the lives of patients and their families, as well as society (Ra et al., 2018).
Children and adolescents with ADHD are more likely to experience a variety of adverse
outcomes compared to those without ADHD (Danielson et al., 2018). These consequences
include lower academic attainment, impaired social functioning, substance use and gambling,
and lead to low income and lower quality of life in adulthood (Danielson et al., 2018; Fleming et
al., 2017; Fletcher, 2014; Groenman, Janssen, & Oosterlaan, 2017; Molina et al., 2013; Ros &
Graziano, 2017). Thus, it is crucial that early identification and earlier interventions of ADHD
among at-risk populations may alter the trajectory and long-term consequences this disorder
(Halperin et al., 2012).
Prevalence of Digital Media Use of Adolescents
Meanwhile, recently the use of digital media has dramatically increased, most notably
among adolescent populations across the world (Lemola, Perkinson-Gloor, Brand, Dewald-
Kaufmann, & Grob, 2015; Romer & Moreno, 2017). For instance, 87% of U.S adolescents
reported owning or having access to a mobile phone, tablet, desk or laptop computer, or
videogame console (Lenhart et al., 2015). Smartphones and other portable mobile devices
provide ubiquitous access to numerous modern digital media activities that are highly popular
Ra, Dissertation
46

46
amongst youth such as social networking, streaming movies or music, texting, video chatting,
and mobile videogames (Lenhart et al., 2015; Ra et al., 2018). Approximately 90% of American
teens reported being active users of social media (Sherman, 2016). There is evidence that modern
digital media exposure activates neural circuits underlying attention and impulse control
(Sherman, Payton, Hernandez, Greenfield, & Dapretto, 2016). Thus, there has been growing
public health concern that increasing digital media usage may adversely impact population
neurobehavioral health such as ADHD (George & Odgers, 2015).
Possible Mechanisms Linking Digital Media Use with ADHD
As aforementioned, exposure to modern digital media robustly activates neurocircuitry
involved in attention and impulse control (Sherman et al., 2016) and could confer risk for ADHD.
Excessive digital media use may promote an attentional style of rapid scanning and shifting that
inhibits ability to sustain focus (Nikken & de Haan, 2015). Use of modern digital media could
also habituate youth to immediate feedback and stimulation, and in turn stunt the development of
impulse control (Nikken & de Haan, 2015). Use of older digital media platforms, such as
television viewing and console video game playing, have been linked with ADHD (Nikken & de
Haan, 2015). However, whether ADHD is associated with use of modern digital media is unclear.
Digital Media Use and Urgency
Urgency may be related to digital media use and ADHD in adolescents. Positive urgency
and negative urgency are affect traits (Cyders & Smith, 2008) where positive urgency (PU)
refers to the tendency to engage in rash actions in response to positive affective states whilst
negative urgency (NU) refers to the tendency to engage in rash actions in response to affective
states (Cyders & Smith, 2008; Marmorstein, 2013; Pedersen et al., 2016). In previous studies, PU
Ra, Dissertation
47

47
and NU were tested as an aggregated construct of impulsivity as they are included in a multi-
facet construct of impulsivity (lack of planning, lack of perseverance, sensational seeking,
positive urgency and negative urgency) (Lynam, Smith, Cyders, Fischer, & Whiteside, 2007). It
has been well established that the level of impulsivity is related to the severity of Internet
addiction (Lee et al., 2012), problematic internet use (Mottram & Fleming, 2009), and
compulsive internet use (Meerkerk, van den Eijnden, Franken, & Garretsen, 2010). In addition, a
few studies found that impulsivity is positively related to increased time of internet game or
video game use (Kwon, 2001; Park, Han, Kim, Cheong, & Lee, 2016).
Although it is relatively well established that impulsivity is positively related to increased
use of digital media, its relation to the specific facet of impulsivity such as urgency remains
poorly investigated. Thus, investigation of the unique, specific contribution of each of these
facets of impulsivity to mental health diagnosis such as ADHD has been strongly recommended
(G. T. Smith & Cyders, 2007; Whiteside & Lynam, 2001). Among these various dimensions of
impulsivity, urgency has recently received increased attention (Cyders & Smith, 2008). However,
to my knowledge, there are only two studies that have investigated the relationships between
DMU and PU/NU. Billieux and his colleagues (2007) found that negative urgency (NU) was
positively correlated with increased use of the cellular phone, and NU was the most important
predictor of perceived dependence on the cellular phone. They suggested that people with high
NU tend to use their mobile phones more often with a greater feeling of dependence due to
feeling compelled to provide for their needs as soon as possible (Joel Billieux, Van der Linden,
d'Acremont, Ceschi, & Zermatten, 2007). Another research from the same team was conducted
in the following year (Joë l Billieux, Van der Linden, & Rochat, 2008), and reported that NU may
be the strongest predictor of problematic use of the mobile phone. They concluded that people
Ra, Dissertation
48

48
with high-negative-urgency tend to use their mobile phone more often because they feel
compelled to provide for their needs as soon as possible (Joë l Billieux et al., 2008). In addition,
these people are more likely to use their mobile phone to satisfy certain strong impulses to
relieve negative affect instantly (Joë l Billieux et al., 2008).
Yet, these two studies are limited to mobile phone use only rather than an integrated
measure of different digital media such as video games or Internet use. To summarize, there is
limited number of research on the relationship between digital media use and a specific facet of
urgency (PU/NU). Moreover, to the best of my knowledge, neither PU nor NU has been
examined as a moderator between digital media usage and ADHD in previous research. Given
previous findings, understanding both positive and negative urgency in relation to digital media
use and ADHD may provide new insight to ADHD prevention and intervention programs.
Specific Aims and Hypotheses
To address the present research gaps in this topic, this study aimed to examine 1) the
association between the baseline use of modern digital media and subsequent likelihood of
meeting symptom criteria for ADHD, and 2) the moderating effects of positive urgency (PU)
and negative urgency (NU) between the association of the two across a 24-month follow-up
period in high school students in Los Angeles who did not meet symptom criteria for ADHD at
baseline. The basic conceptual model is displayed in Figure 3.1.
Ra, Dissertation
49

49

Figure 3.1. Conceptual framework of Study 2

Aim 4: To test whether digital media use (DMU) predicts increased risk of ADHD symptoms.
Hypothesis 7: DMU may be positively associated to adolescent’s ADHD symptoms.
Aim 5: To test whether PU/NU moderates the association between DMU and ADHD.
Hypothesis 8: The association DMU with adolescents’ Symptom status may be
moderated by PU (i.e. adolescents with higher PU may be at higher risk of having
symptom status when the engagement of DMU is constant).
Hypothesis 9: The association DMU with adolescents’ Symptom status may be
moderated by NU (i.e. adolescents with higher NU may be at higher risk of having
symptom status when the engagement of DMU is constant).

Ra, Dissertation
50

50
METHODS
Participants and Procedures
This study used data from the Happiness & Health (H&H) Study, a longitudinal survey of
health behaviors and mental health amongst students enrolled in ten public high schools in Los
Angeles, CA, USA (Leventhal et al., 2015). Among 40 public schools that were reached to
participate based on their representation of diverse demographic characteristics and proximity,
10 schools agreed to participate. All 9
th
grade students who were enrolled in standard
educational programming at the 10 schools in Fall 2013 were eligible to participate. For study
enrollment, students and their parents were required to provide written or verbal assent and
consent, respectively. Study enrollees completed semiannual surveys beginning in Fall 2013
until Fall 2016. This analysis used the 3
rd
to 7
th
survey waves, which was when measures for this
report were introduced into the study.
As presented in Figure 2.2, of the 4,100 eligible 9
th
grade students, 3,396 students and
their parents provided active written or verbal assent and consent, respectively, and enrolled in
the cohort. The variables in this report were first measured at Fall of 10
th
grade (baseline for the
current study; N surveyed=3,051, 98.7%), and our analytic sample included 2,508 students who
completed key measures (the digital media measure, the ADHD measure, the urgency measure,
and the MVPA measure) and did not meet Diagnostic and Statistical Manual of Mental
Disorders – 4
th
Edition (DSM-IV) (Association, 2013) symptom criteria for ADHD at baseline.
Data analyzed involved five semiannual assessments: baseline (Fall 2014, 10
th
grade; N=2,508),
6-month (Spring 2015, 10
th
grade; N=2,402), 12-month (Fall 2015, 11
th
grade; N=2,293), 18-
month (Spring 2016, 11
th
grade; N=2,179), and 24-month (Fall 2016, 12
th
grade; N=2,234)
follow-ups. At each assessment, paper-and-pencil surveys were administered in students’
Ra, Dissertation
51

51
classrooms. Students not present day of data collections were offered abbreviated telephone or
online surveys which omitted the digital media use and ADHD questionnaires, and are excluded
from analysis. This study was approved by the University of Southern California Institutional
Review Board.

Ra, Dissertation
52

52
Figure 3.2. Study accrual flow chart
4,100 Eligible students
186 Excluded for analyses due to positive ADHD status at baseline
3,874 Provided assent
3,051 Administered full wave 3 survey (Baseline)

345 Did not provide data at wave 3 assessment
230 Completed abbreviated survey omitting ADHD and digital media use measures
106 Lost to follow-up
9 Declined to participate

226 Did not provide assent
3,396 Enrolled at the time of the parent study wave 1 survey
478 Did not receive parental consent
439 Consent declined by parent
39 Did not return consent form or parent unreachable

2,508 Analytic sample with ADHD data >1 follow-up
2,402 Data available at 6-month follow-up
2,293 Data available at 12-month follow-up
2,179 Data available at 18-month follow-up
2,234 Data available at 24-month follow-up

343 Excluded (incomplete data on key variables)
208 Did not complete digital media measure and ADHD measure at baseline
83 Did not complete urgency measure and physical activity measure at baseline
66 Did not complete 6-, 12-, 18- or 24- month follow-ups

2,694 Completed ADHD, digital media and urgency measures at baseline (W3) and follow-ups (W4-7)
Ra, Dissertation
53

53
Measures
Past 6-Month ADHD Symptom Status
At each assessment, students were administered the current ADHD symptom scale
(Barkley, 1991), which contains 18 questions aligned with each of the DSM-IV (Association,
2013) ADHD symptom criteria: 9 items of inattention symptoms (e.g., difficulty organizing and
completing tasks) and 9 items of hyperactivity/impulsivity symptoms (e.g., difficulties remaining
still or with task persistence). Respondents rated how frequently they experienced each
individual symptom during the past 6 months on a 4-point Likert scale (Never or Rare,
Sometimes, Often, Very Often). Consistent DSM-IV guidelines (Association, 2013), adolescents
who reported experiencing “Often” or “Very Often” for >6 inattention symptoms or >6
hyperactivity/impulsivity symptoms were classified as meeting symptom criteria for ADHD
within the prior 6-month interval. They were dichotomized “having ADHD” vs. “not having
ADHD” for waves 3-7 (follow-ups).
Modern Digital Media Use
At baseline, participants were administered a checklist instructing students to select one
of four options to indicate their typical frequency of engaging in 14 digital media activities in the
past week (0 times [none]; 1-2 times a week [low]; 1-2 times per day [moderate]; many times per
day [high]); the activities are listed in Figure 2.3. Responses were categorized into binary options
to distinguish high frequency (many times per day) vs. low frequency (all other responses). The
total number of digital media activities engaged at a high frequency rate were summed to
cumulative digital media use frequency index (range: 0-14), which was the primary exposure
variable.

Ra, Dissertation
54

54
Digital media activities
1. Checking social media sites
2. Posting own photos, images, videos, status updates or blogs
3. Liking or commenting on other’s statuses, wall posts or pictures
4. Sharing other's photos, images, videos, status updates, blogs, articles or news
5. Browsing or viewing images or videos
6. Reading online blogs, articles, news, forums or books
7. Streaming television or movies
8. Streaming or downloading music
9. Chatting online
10. Texting
11. Video-chatting
12. Online shopping or browsing
13. Playing games with friends or family on a console, computer or smartphone
14. Playing games by yourself on a console, computer or smartphone

Figure 3.3. 14 items included in digital media activities

Positive Urgency and Negative Urgency
The UPPS-P (Urgency (negative), Premeditation, Perseverance, Sensation seeking, and
Positive urgency) Impulsive Scale (Lynam et al., 2007) was used to measure five facets of
impulsivity including positive urgency and negative urgency. The 59-item scale developed by
Cyders and Smith (Cyders & Smith, 2007) included 12 negative urgency items (e.g., I often act
without thinking when I am upset) and 14 positive urgency items (e.g., I tend to lose control
when I am in a great mood). The items were randomly ordered, and each item is rated on a four-
point Likert scale ranging either from 1 (strongly agree) to 4 (strongly disagree) or 1 (strongly
disagree) to 4 (strongly agree). Since the items from different scales run in different directions,
some items were reverse-scored. That is, all of the scales run in the direction meaning that higher
scores indicate more impulsive behavior (Cyders & Smith, 2007). The mean of the available
items (participants answered at least 70% of the items) was calculated for Positive Urgency (PU)
and Negative Urgency (NU) (Cyders & Smith, 2007). The Cronbach’s alphas of the two scales
were .96 for PU and .92 for NU (See appendix II for more details).
Ra, Dissertation
55

55
Covariates
To examine whether associations between digital media use and Symptom status were
independent of potential confounding influences, baseline factors associated with digital media
use and/or ADHD were selected as covariates (Connor et al., 2003; Sibley et al., 2011; Yen, Ko,
Yen, Wu, & Yang, 2007). Sociodemographic covariates included age; gender; race/ethnicity; and
subsidized lunch eligibility. Subsidized lunch eligibility was based in part on the student’s family
income level relative to the U.S. federal poverty guideline (no free lunch [income >185%
poverty line], free or reduced lunch [income <130% - 185% above poverty line]).
Sociodemographic covariates were assessed with investigator-defined forced-choice items (Table
3.1).
Potential confounders
Given frequent behavioral and emotional problems among youth with ADHD (Elia,
Ambrosini, & Berrettini, 2008), additional covariates were included in the adjusted models.
Baseline delinquent behavior was measured with a sum of frequency ratings for engaging in 11
different behaviors and conduct disorder problems within the past 6 months (e.g. “ stealing” and
“skipping school”; range 1 [never] to 6 [10 or more times]; Cronbach’s alpha internal
consistency estimate = .73) (Thompson, Ho, & Kingree, 2007). Past-week depressive symptoms
were measured using the Center for Epidemiological Studies Depression Scale (CESD) (Radloff,
1991) providing sum score for 20 symptoms (0=rarely or none of the time to 3=most or all of
the time; alpha = .82). Self-reported substance use were operationalized as a three-level variable
(current [past 30 days] vs. past [ever use, but no use in the past 30 days] vs. never use of
cigarettes, alcohol, or cannabis), (Leventhal et al., 2015). Lastly, family history of substance use
Ra, Dissertation
56

56
(i.e., having siblings, parents, grandparents who smoked cigarettes or had an alcohol or drug
abuse problem; yes/no) were included.
Statistical Analysis
For Study Aim 3 , given the potential for reciprocal pathways (or selection effects) in
which youth with ADHD may be more liable to seek out digital media usage (Nikkelen,
Valkenburg, Huizinga, & Bushman, 2014), only adolescents with negative baseline ADHD
status (having no ADHD) were included in the primary analyses. Primary analyses used
generalized estimating equations (GEEs) (Zeger, Liang, & Albert, 1988), utilizing repeated
measures binary (link logit) logistic regressions for dichotomous outcomes in which each
participant had up to four time points of follow-up data (at 6-, 12-, 18-, and 24-months) modeled
using an exchangeable correlation structure (Hubbard et al., 2010). GEEs model the odds of
positive (vs. negative) ADHD symptom status at the follow-ups based on baseline digital media
use regressor values. The set of models used the digital media activities composite score as the
primary regressor. Models were tested with time (6-, 12-, 18-, and 24-month follow-up) and
school as fixed effects and then retested after they are adjusted for baseline covariates (sex, age,
ethnicity, subsidized lunch status, substance use, delinquent behavior and CESD for depressive
symptoms). The digital media use regressor estimate indicates the averaged association with
ADHD collapsed across the four follow-ups. Digital media use (DMU)  Time interaction terms
were added in subsequent models to test whether the association was changed across follow-up
time points. For Study Aim 4, same analyses were conducted including the interaction term of
DMU  PU (or NU) to test whether the association between DMU and ADHD symptom status is
changed by the level of PU (or NU).
Ra, Dissertation
57

57
Since only participants with complete digital media use and ADHD data were included in
the analysis, missing data for the adjustment covariates was permitted (see Table 3.1 for
available data for each covariate) and accounted by multiple-imputation (Rubin, 2004), which
replaces each missing value with a set of plausible values that represent the uncertainty about the
correct value to impute. Using the Markov-chain Monte Carlo method for missing at random
assumptions and the available covariate data, five multiple-imputed data sets were created. The
parameter estimates from the models tested in each imputed data set were pooled into a single
estimate. Continuous variables were standardized (M=0, SD=1) to facilitate interpretation of
parameter estimates across covariates with different scaling. Statistical analyses were conducted
using SPSS version 24 (IBM Corp., Armonk, NY). Results were reported as odds ratios (ORs)
with 95% Confidence intervals (CIs). Significance was set to 0.05 and tests were 2-tailed.
RESULTS
Study Sample
Participant accrual, sample size, and reasons for exclusion for this report’s analytic
sample are depicted in figure 3.2. Among all 4,100 9
th
grade students eligible for the parent study,
3,874 assented (94.5%), for whom 3,396 parents (87.7%) provided consent and enrolled. Of the
3,051 participants who were administered the baseline (wave 3) full-length survey, 343 were
excluded due to incomplete data on digital media use or ADHD measures. After excluding 186
participants who met symptom criteria for ADHD at baseline, this report’s primary analytic
sample is 2,508 (See Table 3.1 for more details).

Ra, Dissertation
58

58
Descriptive Analyses
Table 3.1 reports descriptive statistics for digital media use and other baseline
characteristics for the analytic sample of participants with negative baseline ADHD status
(N=2,508). There were slightly more than females than males (54.7% vs. 45.3%) in the analytic
sample, and the mean age of the participants was 15.5 years old. Hispanic participants was the
major ethnicity and constituted 46.3%. Of 2,274 participants, 48.1% received free or reduced
lunch. Over a half of the participants (53.4%) responded they had never used any substance
including cigarette. Of 2,399 participants, 60.2% reported they had not had a family history of
substance use (i.e. having siblings, parents, grandparents who smoked cigarettes or had an
alcohol or drug abuse problem; yes/no).

Ra, Dissertation
59

59
Table 3.1. Descriptive Statistics of Sample Characteristics

TOTAL

(N=2,508)
Variables N (%) or Mean (SD)
Gender (n=2,508)
a

Male 1137(45.3)
Female 1371(54.7)
Age, (n=2,442)
a
15.50(0.5)
Race/ethnicity (n=2,450)
a

Hispanic 1160(47.3)
Asian 430(17.6)
Black/African American 105(4.3)
White 374(15.3)
Other 381(15.6)
Subsidized lunch eligibility status (n=2,274)
a

No free lunch 1181(51.9)
Free or reduced lunch 1093(48.1)
Substance use (n=2,489)
a

Never 1330(53.4)
Past user 588 (23.6)
Current user 571(22.9)
Family history of substance use (n=2,399)
a

Ra, Dissertation
60

60
Note.
a
The denominator is for the data in column 1 and is provided due to missing data for this variable.
b
Score
range: 1 (never) to 6 (10 or more) within the past 6 months for each item × 11 items.
c
Score range: 0 (rarely or none
of the time; 0-1 day) to 3 (most or all of the time; 5-7 days) for each symptom × 20 symptoms.
d
Score range: 1
(frequent use; many times a day) for each digital media activity in the past week × 20 activities.
e
Score range: 1
(strongly agree) to 4 (strongly disagree) for the mean of 14 items.
f
Score range: 1 (strongly agree) to 4 (strongly
disagree) for the mean of 12 items.
g
Score range: 0 (none) to 3(very often) for each symptom x two 9-item subscales
(inattentive amd hyperactive/impulsive subtypes). Abbreviations: CESD = Center for Epidemiologic Studies
Depression Scale; N = Sample size; SD = Standard Deviation.

Yes 954(39.8)
No 1445(60.2)
Delinquent behavior (n=2,508)
b
14.28(4.2)
CESD scale for depressive symptoms (n=2,482)
a,c
14.40(12.1)
Digital media use frequency
d
3.61(3.28)
Positive Urgency
e
1.48(0.6)
Negative Urgency
f
1.89(0.7)
ADHD symptom score at follow-up
a,g

6-month (n=2,406) 12.53(9.4)
12-month (n=2,294) 10.19(9.1)
18-month (n=2,182) 10.58(9.4)
24-month (n=2,231) 10.74(9.1)
Ra, Dissertation
61

61

Figure 3.4. ADHD prevalence at each follow-ups (n=2,508)

Figure 3.5. ADHD mean prevalence across all follow-ups (n=2,508)

Figure 3.5 depicts the relationship between the digital media use frequency and ADHD
occurrence across follow-ups. There is a modest positive relationship between digital media use
frequency and the occurrence of ADHD symptoms at 2-year-follow-ups.
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Digital Media Use Frequency
6-month follow-up
12-month follow-up
18-month follow-up
24-month follow-up
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
14.0%
16.0%
18.0%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Digital Media Use Frequency
Ra, Dissertation
62

62
Associations Between Baseline Modern Digital Media Use and ADHD Status at Follow-Up
among Those with Negative ADHD Status at Baseline
Among youth negative for ADHD at baseline, 164 (6.8%), 110 (4.8%), 125 (5.7%) and
126 (5.6%) transitioned to positive ADHD status at 6-, 12-, 18-, and 24-month follow-up
respectively. The unadjusted estimate for the association of the digital media use frequency at
baseline with positive ADHD status at follow-up was statistically significant (OR, 1.44; 95% CI,
1.21-1.71). The associations remained after adjustment for baseline sociodemographic
covariates, substance use, depressive symptoms, and delinquent behavior (OR, 1.39; 95% CI,
1.15-1.67). Each one standard deviation unit increase in digital media use frequency was
associated with a 39% (95% CI, 1.09-1.41) increase in odds of transitioning to positive ADHD
status (vs. maintaining negative ADHD status) at follow-up. The interaction term between the
digital media frequency and time was not significant (p=.147), meaning that the association
between the digital media use and ADHD status at follow-ups did not change across follow-ups
(time). In this model, gender was the only sociodemographic covariate associated with follow-
up ADHD status (OR, 0.52; 95% CI, 0.41-0.66), wherein girls were less likely to transition to
positive ADHD symptom status at follow-ups than boys. Additionally, adolescents with higher
baseline depressive symptoms and more delinquency were more likely to transition to positive
ADHD status at follow-up (OR, 1.22 [95% CI, 1.08-1.38]; OR, 1.58 [95% CI, 1.41-1.78]
respectively) (Table 3.2).

Ra, Dissertation
63

63
Table 3.2. Association of baseline digital media use with ADHD status across 6- 12- 18- and
24-month follow-ups among ADHD-negative adolescents

Positive (vs. Negative) ADHD Status at
Follow-up
Baseline Regressors OR (95% CI)
a
P Value
Unadjusted Model
b

Digital media use frequency
c,d
1.44 (1.21, 1.71)

<.001
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.90, 1.04) 0.324
Digital media use frequency × time
e
0.95 (0.90, 1.01) 0.125
Adjusted Model
f

Digital media use frequency
c,d
1.39 (1.15, 1.67)

<.001
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.93, 1.00) 0.371
Digital media use frequency × time
e
0.95 (0.90, 1.02) 0.147
Female (vs. male) gender 0.52 (0.41, 0.66) <.001
Age
d
0.96 (0.88, 1.05)

0.343
Race/ethnicity
Hispanic Ref
Asian 0.88 (0.55, 1.39) 0.563
Black/African American 1.34 (0.76, 2.39)

0.309
White 0.97 (0.64, 1.49)

0.953
Other 1.06 (0.74, 1.51) 0.659
Subsidized lunch eligibility status (yes vs. no) 0.79 (0.58, 1.08) 0.138
Substance use
Never Ref
Past 1.14 (0.86, 1.52) 0.355
Current 1.21 (0.89, 1.66) 0.229
Family history of substance use (yes vs. no) 1.09 (0.84, 1.40) 0.523
Delinquent behavior
d
1.22 (1.08, 1.38)

<.001
CESD scale for depressive symptoms
d
1.58 (1.41, 1.78)

<.001
Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; OR odds ratio, CI = Confidence Interval.
Note.
a
OR(95% CI) estimates and corresponding p-values from respective regressor parameter estimates for
association with follow-up ADHD status from repeated binary logistic generalized estimating equations, including
school fixed effects.
b
Model included digital media use as primary regressor.
c
Cumulative index of binary frequent
digital media use (Range = 0 – 14).
d
Rescaled (M=0, SD=1) such that the ORs indicate the change in odds in the
outcome associated with an increase in 1 SD unit on the covariate continuous scale.
e
Interaction term added in
subsequent model; OR’s for other regressors or covariates are from the model excluding the interaction term.
f
Model
included digital media use and other covariates as simultaneous regressors.

Ra, Dissertation
64

64
Moderating Effects of Positive Urgency in the Association Between Baseline Modern
Digital Media Use and ADHD Status at Follow-Ups Among ADHD-Negative Adolescents at
Baseline
The unadjusted estimate for the association of the digital media use frequency at baseline
with positive ADHD status at follow-up and Positive Urgency (PU) were statistically significant
(OR, 1.39 [95% CI, 1.16-1.67]; OR, 1.71 [95% CI, 1.53-1.091] respectively). The associations
remained after adjustment for baseline sociodemographic covariates, substance use, depressive
symptoms, and delinquent behavior (OR, 1.37 [95% CI, 1.14-1.65]; OR, 1.53 [95% CI, 1.35-1.73]
respectively). Each one standard deviation unit increase in PU was associated with a 53% (95%
CI, 1.35-1.73) increase in odds of transitioning to positive ADHD status (vs. maintaining
negative ADHD status) at follow-up. However, the interaction term of digital media use
frequency and Positive Urgency (PU) were not statistically significant in both unadjusted and
adjusted models (p=.376, .836 respectively) (Table 3.3).

Ra, Dissertation
65

65
Table 3.3. Moderating effects of positive urgency in the association of baseline digital media
use with ADHD status across 6- 12- 18- and 24-month follow-ups among ADHD-negative
adolescents

Positive (vs. Negative) ADHD Status at
Follow-up
Baseline Regressors OR (95% CI)
a
P Value
Unadjusted Model
b

Digital media use frequency
c,d
1.39 (1.16, 1.67) <.001
Positive Urgency 1.71 (1.53, 1.91) <.001
Digital media use frequency × positive urgency
e
0.96 (0.88, 1.05) 0.376
Time (6- 12- 18- and 24-month follow-ups) 0.96 (0.90, 1.04) 0.315
Digital media use frequency × time
e
0.95 (0.90, 1.01) 0.128
Adjusted Model
f

Digital media use frequency
c,d
1.37 (1.14, 1.65)

<.001
Positive Urgency 1.53 (1.35, 1.73) <.001
Digital media use frequency × positive urgency
e
0.99 (0.89, 1.10) 0.836
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.93, 1.00) 0.348
Digital media use frequency × time
e
0.95 (0.89, 1.02) 0.14
Female (vs. male) gender 0.54 (0.42, 0.68) <.001
Age
d
0.96 (0.88, 1.04)

0.321
Race/ethnicity

Hispanic Ref
Asian 0.79 (0.50, 1.25) 0.316
Black/African American 1.32 (0.73, 2.40) 0.361
White 0.96 (0.63, 1.48) 0.861
Other 1.02 (0.72, 1.46)

0.896
Subsidized lunch eligibility status (yes vs. no) 0.79 (0.57, 1.08)

0.134
Substance use
Never Ref
Past 1.06 (0.80, 1.41) 0.672
Current 1.12 (0.81, 1.54) 0.497
Family history of substance use (yes vs. no) 1.03 (0.80, 1.32) 0.838
Delinquent behavior
d
1.16 (1.00, 1.36)

0.055
CESD scale for depressive symptoms
d
1.48 (1.30, 1.68)

<.001
Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; OR odds ratio, CI = Confidence Interval.
Note.
a
OR(95% CI) estimates and corresponding p-values from respective regressor parameter estimates for association with
follow-up ADHD status from repeated binary logistic generalized estimating equations, including school fixed effects.
b
Model
included digital media use as primary regressor.
c
Cumulative index of binary frequent digital media use (Range = 0 – 14).
d
Rescaled (M=0, SD=1) such that the ORs indicate the change in odds in the outcome associated with an increase in 1 SD unit on
the covariate continuous scale.
e
Interaction term added in subsequent models; OR’s for other regressors or covariates are from the
model excluding the interaction terms.
f
Model included digital media use and other covariates as simultaneous regressors.
Ra, Dissertation
66

66
Moderating Effects of Negative Urgency in the Associations Between Baseline Modern
Digital Media Use and ADHD Status at Follow-Ups among ADHD-Negative Adolescents at
Baseline
The unadjusted estimate for the association of the digital media use frequency at baseline
with positive ADHD status at follow-up and Negative Urgency (NU) were statistically
significant (OR, 1.38 [95% CI, 1.14-1.67]; OR, 1.72 [95% CI, 1.52-1.94] respectively). The
associations remained after adjustment for baseline sociodemographic covariates, substance use,
depressive symptoms, and delinquent behavior (OR, 1.34 [95% CI, 1.11-1.62]; OR, 1.54 [95%
CI, 1.33-1.78] respectively). Each one standard deviation unit increase in NU was associated
with a 54% (95% CI, 1.33-1.78) increase in odds of transitioning to positive ADHD status (vs.
maintaining negative ADHD status) at follow-up. However, the interaction term of digital media
use frequency and NU were not statistically significant in both unadjusted and adjusted models
(p=.638, .469 respectively) (Table 3.4).

Ra, Dissertation
67

67
Table 3.4. Moderating effects of negative urgency in the association of baseline digital
media use with ADHD status across 6- 12- 18- and 24-month follow-ups among ADHD-
negative adolescents

Positive (vs. Negative) ADHD Status at
Follow-up
Baseline Regressors OR (95% CI)
a
P Value
Unadjusted Model
b

Digital media use frequency
c,d
1.38 (1.14, 1.67) <.001
Negative Urgency 1.72 (1.52, 1.94) <.001
Digital media use frequency × Negative urgency
e
1.03 (0.91, 1.16) 0.638
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.90, 1.04) 0.339
Digital media use frequency × time
e
0.95 (0.89, 1.02) 0.163
Adjusted Model
f

Digital media use frequency
c,d

1.34 (1.11, 1.62)
<.001
Negative Urgency 1.54 (1.33, 1.78) <.001
Digital media use frequency × Negative urgency
e
1.04 (0.93, 1.18) 0.469
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.90, 1.04) 0.368
Digital media use frequency × time
e
0.96 (0.89, 1.02) 0.175
Female (vs. male) gender 0.48 (0.38, 0.61) <.001
Age
d
0.95 (0.87, 1.04) 0.236
Race/ethnicity
Hispanic Ref
Asian 0.80 (0.50, 1.27) 0.341
Black/African American 1.42 (0.80, 2.49) 0.227
White 0.97 (0.64, 1.49) 0.907
Other 1.03 (0.72, 1.46) 0.884
Subsidized lunch eligibility status (yes vs. no) 0.79 (0.58, 1.08) 0.139
Substance use
Never Ref
Past 1.09 (0.82, 1.44) 0.57
Current 1.17 (0.85, 1.61) 0.327
Family history of substance use (yes vs. no) 0.98 (0.76, 1.26) 0.865
Delinquent behavior
d
1.15 (1.00, 1.33) 0.051
CESD scale for depressive symptoms
d
1.34 (1.17, 1.53) <.001
Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; OR odds ratio, CI = Confidence Interval.
Note.
a
OR(95% CI) estimates and corresponding p-values from respective regressor parameter estimates for association with
follow-up ADHD status from repeated binary logistic generalized estimating equations, including school fixed effects.
b
Model
included digital media use as primary regressor.
c
Cumulative index of binary frequent digital media use (Range = 0 – 14).
d
Rescaled (M=0, SD=1) such that the ORs indicate the change in odds in the outcome associated with an increase in 1 SD unit on
the covariate continuous scale.
e
Interaction term added in subsequent models; OR’s for other regressors or covariates are from the
model excluding the interaction terms.
f
Model included digital media use and other covariates as simultaneous regressors.
Ra, Dissertation
68

68
DISCUSSION
Despite concerns regarding the impact of increasing digital media usage on
neurobehavioral development (George & Odgers, 2015), there is remarkably little rigorous data
addressing the public health implications of youth modern digital media use. To the best of my
knowledge, this is one of the first studies to assess associations between digital media use and
ADHD, which is a condition associated with substantial adverse educational, social, legal, health
and behavioral outcomes (Brook, Brook, Zhang, Seltzer, & Finch, 2013; Nigg, 2013) and
increased mortality (Dalsgaard, Østergaard, Leckman, Mortensen, & Pedersen, 2015).
The results of this study demonstrate that frequent use of various forms of modern digital
media was associated with increased odds of meeting symptom criteria for ADHD over a 24-
month follow-up of adolescence in 2,508 students who were not positive for ADHD at baseline.
Moreover, both positive and negative urgency were positively associated with ADHD symptom
criteria, however, neither positive nor negative urgency moderated the impact of digital media
use frequency on ADHD symptom occurrence among those adolescents.
Digital Media Use and ADHD Symptom Criteria
These findings complement and advance existing evidence. A meta-analysis involving 45
samples of more than 150,000 youth collected during or prior to 2011 found that older forms of
digital media use in (e.g., television viewing and videogame playing) was consistently associated
with ADHD and related outcomes (Nikkelen et al., 2014). Also, a cross-sectional study of
Chinese youth reported associations between mobile phone use and inattention (Zheng et al.,
2014). A study high-risk 11 to 15 year olds found that on days youth used modern forms of
digital media more often, they experienced more severe ADHD symptoms (George, Russell,
Piontak, & Odgers, 2017). Furthermore, the current study reports new longitudinal evidence of
Ra, Dissertation
69

69
an association between use of modern digital media platforms and ADHD across a key
developmental period.
Positive and Negative Urgency and ADHD Symptom Criteria
The findings also indicate that both positive and negative urgency were positively
associated with ADHD symptom criteria, however, neither of those moderated the impact of
digital media use frequency on ADHD symptom occurrence among those adolescents. These
findings of the main effects of PU and NU on ADHD are consistent with previous literature.
Wiklund and his colleagues (2017) conducted a case-control study with 102 youth and found a
positive link between negative urgency and ADHD symptoms (Wiklund, Yu, Tucker, & Marino,
2017). Another study using 720 high school students in France reported that a high urgency score
(both PU and NU) can be a risk factor for developing ADHD (Romo et al., 2015). According to
Billieux et al. (2010), both PU and NU have been related to a wide range of maladaptive
behaviors related to ADHD, and the team suggested that urgency-related behaviors may explain
ADHD symptoms based on the results from their cross-sectional study (Joë l Billieux, Gay,
Rochat, & Van der Linden, 2010).
Urgency is related to emotional instability, which is a facet of affect traits (Letzring &
Adamcik, 2015). Adolescents higher on urgency are more sensitive to either positive or negative
cues and are more likely to identify certain conditions as threatening (Letzring & Adamcik,
2015). They experience extreme positive or negative affective states more frequently and more
strongly than others (Joë l Billieux et al., 2010). Those may reflect a disposition toward ADHD
symptoms as youth with ADHD are often characterized by extreme affect like positive affect,
negative affect and affect volatility (Martel & Nigg, 2006; Nigg et al., 2002a, 2002b). Thus,
people high on urgency may be prone to experience ADHD symptoms.
Ra, Dissertation
70

70
Although we found that both positive and negative urgency were mainly associated with
ADHD symptom criteria, neither moderated the association between the digital media use
frequency and ADHD symptom criteria. The initial hypotheses was drawn from previous
research that there is a positive relationship between urgency and digital media use such as
smartphone use (Joë l Billieux et al., 2010; Joel Billieux et al., 2007; Joë l Billieux et al., 2008). In
addition, a few previous studies reported the positive relationship between both PU and NU and
ADHD (Wiklund, Yu, Tucker, & Marino, 2017; Romo et al, 2015; (Davis-Becker et al., 2014).
Unlike hypothesized, the results did not support the moderating effects of positive or negative
urgency in the association between digital media use and ADHD. Based on the preliminary
analyses, those three variables are highly correlated to each other, which is consistent with
previous findings. There are possibilities that urgency may be a mediator, not a moderator. That
is, either positive or negative urgency impacts adolescents’ digital media usage and in return, that
increases the risk of having ADHD symptom status. Thus, in future research, the mediation
model needs to be tested to better understand the dynamic relationship between those three.

Limitations
Although this study provides valuable insights into associations between modern digital
media use and ADHD, it is not without limitations. First, a potential limitation of the current
study lies in assessing ADHD symptom criteria. we used Current Symptoms Scale-Self Report
Form (Barkley, 1991) to assess ADHD symptom status. Responses on the rating scale are not
sufficient to render a diagnosis of ADHD, but may be consistent with the disorder. However,
further clinical evaluation is necessary (Harpin, V. 2017). Second, for digital media use, high
frequency digital media use was operationalized as “multiple times per day,” which may be
Ra, Dissertation
71

71
insufficient for distinguishing the highest levels of use common in the adolescent
population.(Lenhart et al., 2015). Additionally, reporting biases (e.g., teens who admit frequent
digital media use may be more willing to report stigmatized conditions, such as ADHD) could
explain why there may be a positive association between modern digital media use and ADHD
(Ra et al., 2018). We used representative samples of high school students in California. Thus, the
generalizability of the results from this study may be limited to the present population and may
not apply to other adolescents. Additional research across a greater diversity of populations and
age ranges to determine generalizability of may be fruitful.
Implications
Despite these limitations, this study has many strengths. First, the exclusion of students
with ADHD at baseline limits reverse causality explanations to examine temporal and causal
inferences between modern digital use and ADHD. The quantitative exposure modeling is
uniquely precise and provided new evidence of a dose-response association, which strengthens
inferences that associations may be causal.
This study is the first to address a controversial question, “Do modern digital media
platforms and the high frequency rate teens use them represent a novel exposure that is
sufficiently neuropathogenic to increase risk of ADHD occurrence?” Modern digital media
platforms are considered more neuropathogenic than traditional digital media such as television
viewing due to their high speed, level of stimulation and ubiquitous use throughout the day. The
2017 special issue in Pediatrics concluded that the evidence to date was insufficient to address
this question because of inadequate assessment of digital media use and poor study designs to
test causal inferences. In addition, this study further explored the moderating effects of positive
Ra, Dissertation
72

72
and negative urgency in the association between modern digital media use and ADHD for the
first time.
Future Research
Further research is needed to determine whether this association is causal. To do so,
clinical interviews to diagnose ADHD and performance based measures of attention and
impulsivity should be included in future research. Additional research across a greater diversity
of populations and age ranges to determine generalizability of may be fruitful. The association
among digital media usage, urgency, and ADHD needs to be carefully examined.
CONCLUSION
In conclusion, amongst Los Angeles area high school students who did not meet
symptom criteria for ADHD, more frequent use of modern digital media was associated with
increased odds of meeting ADHD symptom criteria across a 24-month period of adolescence.
The findings from this study may add new information on digital media use, affects, and
neurobehavioral health of adolescents. Given the increasing global ubiquity of digital media
technologies, careful examination and monitoring of digital media by parents, educators, health
professionals, and policy makers are crucial to protect the neurobehavioral health of the
adolescent population.

Ra, Dissertation
73

73
CHAPTER 4: EXAMINING THE MODERATING EFFECTS OF PHYSICAL
ACTIVITY IN THE ASSOCIATION BETWEEN AFFECTIVE STATES AND ADHD
AND BETWEEN DIGITAL MEDIA USE AND ADHD
INTRODUCTION
Physical Activity and ADHD
In a number of studies, physical activity has been demonstrated to have positive effects
on improving ADHD in children and adolescents (Biddle & Asare, 2011). An intervention study
among boys with ADHD found that engaging in vigorous physical activity has a positive impact
on ADHD symptomatology (Azrin, Ehle, & Beaumont, 2006). A recent study among children
with ADHD also demonstrated that physical activity levels were related to negative affect and
executive functioning (Gawrilow, Stadler, Langguth, Naumann, & Boeck, 2016; Verret, Guay,
Berthiaume, Gardiner, & Bé liveau, 2012). Moreover, Verret and his colleagues (2012) showed
an inverse association between Moderate to Vigorous Physical Activity (MVPA) and ADHD-
related behaviors such as inattention (Verret et al., 2012) in adolescence.
Physical Activity As a Potential Moderator Between Affect and ADHD
As mentioned above, physical activity not only has been shown to have beneficial effects
in improving ADHD symptoms, but also has been demonstrated to improve affective health
(Azrin, Ehle, & Beaumont, 2006; Biddle & Asare, 2011; Gawrilow, Stadler, Langguth, Naumann,
& Boeck, 2016; Verret, Guay, Berthiaume, Gardiner, & Bé liveau, 2012). MVPA has been shown
to acutely increase positive affective states whilst decreasing negative affective states in children
(Genevieve Fridlund Dunton et al., 2014; Genevieve F Dunton et al., 2014; Liao, Shonkoff, &
Dunton, 2015). Physical activity can also alleviate individuals’ affective response to stressors
Ra, Dissertation
74

74
(Hegberg & Tone, 2015; Holmes, 2018; Stults-Kolehmainen & Sinha, 2014). A few studies have
found that trained athletes (vs. untrained individuals) were more likely to exhibit lowered
reactivity to psychological distress (Rimmele et al., 2009; Webb et al., 2013). A similar finding
has been reported among children in a community based study (Spartano, Heffernan, Dumas, &
Gump, 2017). Whilst previous research indicates that physical activity can promote affective
health and improve ADHD symptoms, whether this effect operates primarily through a
moderating effect of MVPA between affects and ADHD remains unexplored.
Physical Activity As a Potential Moderator Between Digital Media Use and ADHD
Most studies indicate that higher rates of digital media use are negatively related to
physical activity (Ballard, Gray, Reilly, & Noggle, 2009; Van Praag, 2008). Physical activity
may be decreasing because people are replacing physical activity with digital media use (DMU)
(Keim, Blanton, & Kretsch, 2004; Nelson, Gortmaker, Subramanian, Cheung, & Wechsler,
2007). For example, video game players may experience a sense of flow or/and lose track of time
(Sherry, 2004). Because playing video games takes up the time that would be spent physically
active, they may be more sedentary. Or, it is possible that two individuals could have the same
level of DMU, but one is also physically active (high MVPA) while the other is very inactive
(low MVPA). Findings of a relationship between digital media use and physical activity are
mixed. Several studies found significant negative associations between digital media use (e.g.
TV screen time, internet use time) and physical activity (Koezuka et al., 2006; Lowry, Wechsler,
Galuska, Fulton, & Kann, 2002; Motl, McAuley, Birnbaum, & Lytle, 2006; Vandewater, Shim,
& Caplovitz, 2004). However, some studies failed to find the relationship between DMU and
physical activity (Hager, 2006; Lanningham-Foster et al., 2006; Laurson, Eisenmann, & Moore,
2008; Roberts & Foehr, 2004; Snoek, van Strien, Janssens, & Engels, 2006). A meta-analysis
Ra, Dissertation
75

75
study on a relationship between digital media use and physical activity in children and
adolescents reported that the relationship is small but negative and suggested that the relationship
cannot be explained using single markers of digital media use such as TV viewing or game use
(Marshall, Biddle, Gorely, Cameron, & Murdey, 2004). Now, DMU has become more diverse
and ubiquitous. Therefore, it is important to investigate a relationship between physical activity
and a measure of integrated digital media use.
This study investigates the role of physical activity in the paths of ADHD – between
affect and ADHD and between DMU and ADHD. The analysis for Aim 5 examines the
moderating effects of physical activity on the association between affective states (the mean
level of PA and NA) and ADHD symptom status among children. The analysis for Aim 6
investigates the moderating effects of physical activity on the relationship between modern
digital media use and ADHD symptom status among adolescents. Aim 5 utilizes data from the
Mothers’ and Their Children’s Health (MATCH) Study (same as Study 1). Aim 6 used data from
the Happiness & Health (H&H) Study (same as Study 2).
Specific Aims and Hypotheses
Aim 5: To test whether MVPA moderates the association of the mean level of affective states
among youth and their ADHD symptoms in the MATCH study.
Hypothesis 8: The association between the mean level of PA and children’s ADHD
symptoms may be moderated by MVPA (i.e. the positive association between the mean
level of PA and ADHD symptom status may be attenuated for children who are more
physically active than the average child).

Ra, Dissertation
76

76
Hypothesis 9: The association between mean level of NA and children’s ADHD
symptoms may be moderated by MVPA (i.e. the positive association between the mean
level of NA and ADHD symptom status may be attenuated for children who are more
physically active than the average child).
Aim 6: To test whether MVPA moderates the association between DMU and ADHD in the H&H
study.
Hypothesis 10: The association DMU with adolescents’ ADHD symptoms may be
moderated by MVPA (i.e. the positive association between DMU and ADHD symptom
status may be attenuated for children who are more physically active than the average
child).

4.1. Conceptual Framework of Study 3, Aim 5 using the MATCH data
Ra, Dissertation
77

77

4.2. Conceptual Framework of Study 3, Aim 6 using the H&H data

METHODS
Participants and Procedures
Data for study Aim 5 were drawn from Wave 1 of the Mothers’ and Their Children’s Health
(MATCH) Study (PI: Dr. Genevieve Dunton). The MATCH Study is a prospective study with a
total of 202 mothers (baseline) and their 8 to 12 year-old children, followed in 6 semi-annual
assessments (e.g., waves) from the fall of 2014 to the spring of 2018. Each wave consists of
accelerometer monitoring; ecological momentary assessment (EMA); and paper-and-pencil
surveys in mothers and children Accelerometers are wearable devices using sensors to detect
body movements in order to estimate the intensity of activity over time (Chen & DAVID R
BASSETT, 2005). EMA is an innovative real-time data collection methodology that uses
electronic devices (e.g., smartphone apps) (Dunton et al., 2015) to minimize participants’ recall
bias and improve external and ecological validity. The MATCH Study is the first of its kind to
Ra, Dissertation
78

78
use EMA to examine the within-day (4 random prompts a day for 8 days) and long-term (6
waves over 3 years) effects of momentary affect on physical activity, and mental health among
mothers and children. Sociodemographic characteristics were assessed through paper-and-pencil
surveys.

Ra, Dissertation
79

79
For Study Aim 6, this study used data from the Happiness & Health Study, a longitudinal
survey of health behaviors and mental health amongst students enrolled in ten public high
schools in Los Angeles, CA, USA (Leventhal et al., 2015). All 9
th
grade students enrolled in
standard educational programming at the 10 schools in fall 2013 were eligible to participate. For
study enrollment, students and their parents were required to provide active written or verbal
assent and consent, respectively. Study enrollees completed semiannual surveys beginning in
Fall 2013. This analysis uses the 3
rd
to 8
th
survey waves, which was when measures for this
report were introduced into the study. At the baseline of this study (Wave 3), 2,712 students
completed key measures and did not meet Diagnostic and Statistical Manual of Mental
Disorders – 4
th
Edition (DSM-IV)(Association, 2013) symptom criteria for ADHD at baseline.
Data analyzed involved five semiannual assessments (see more details in Figure 4.3).

Figure 4.3. Study accrual flow chart

186 Excluded for analyses due to positive ADHD status at baseline

3,051 Administered full baseline (wave 3) survey

2,508 Analytic sample with ADHD data >1 follow-up
2,402 Data available at 6-month follow-up
2,293 Data available at 12-month follow-up
2,179 Data available at 18-month follow-up
2,234 Data available at 24-month follow-up

343 Excluded (incomplete data on key variables)
208 Did not complete digital media measure and ADHD measure at baseline
83 Did not complete urgency measure and physical activity measure at baseline
66 Did not complete 6-, 12-, 18-, or 24-month follow-ups

2,694 Completed ADHD and digital media measures at baseline and follow-up
Ra, Dissertation
80

80
Measures
For Study 3, Aim 5, data from the MATCH study, which was described in Study 1, had been
used to explore the moderating effects of physical activity on the association between affective
states and ADHD.
Physical Activity
Physical activity was measured by the Actigraph, Inc. GT3X model accelerometer across
7 days at wave 1 (Dunton et al., 2015). Participants wore the Actigraph with the adjustable belt
on the right hip at all times except sleeping, bathing, or swimming. The device collected body
movement data in activity counts units for each 30-second epoch. The collected data is assessed
to identify non-wear time (over 60 consecutive minutes of zero activity counts) and valid days
(at least 10 hours of wear time) by Meterplus software (Santech, San Diego, CA). Meterplus
converts activity counts to minutes spent in moderate-to-vigorous physical activity (MVPA). The
cut-off for MVPA is 2,020 activity counts per minutes, which is consistent with national
surveillance studies (Troiano et al., 2008) using age and gender adjusted thresholds for children.
Each child’s total daily MVPA minutes are calculated based on the total minutes of MVPA and
the number of valid days (total daily MVPA minutes = the total minutes of MVPA / the number
of valid days).
ADHD Symptom Status
The outcome, ADHD, was assessed through mothers’ reports on the Child Behavior
Checklist (CBCL) in wave 1. The CBCL is a paper-pencil instrument with 142 questions. It is
the most widely-used standardized measure for evaluating maladaptive behavioral and emotional
problems in youth, and evidence for content, construct, and criterion-related validity is well
Ra, Dissertation
81

81
documented (Achenbach & Edelbrock, 1991). It also serves as a screening tool to identify
ADHD cases in a pediatric primary care referred population (Biederman et al., 2005) since it is
efficient in minimizing the amount of time required by the primary care physician, and is also
cost effective in contrast to the highly intensive works of structured diagnostic interview
methods by a physician (Biederman et al., 2005). Seven items (fails to finish, concentrate, sit still,
impulsive, inattentive, talk much, and loud; see Appendix I for more details) were used to
compute the DSM-V ADHD criteria (below borderline clinical range, within borderline clinical
range and above clinical range) based on gender (boy vs. girl) and age (6-11 vs. 12-18). For the
current analysis, ADHD criteria category was dichotomized into below borderline clinical range
(0) vs. within or above borderline clinical range (1) to refer to the current ADHD symptom status.
Detailed information on use, scoring is accessible at: http://www.aseba.org/.
EMA Affective States
The main predictors, children’s affective states (PA and NA), were obtained from EMA
data from wave 1. Five affective state items were adopted from the Positive and Negative Affect
Schedule for Children (PANAS-C; Ebesutani et al., 2012; Laurent et al., 1999). To assess PA,
two questions were asked, “How 1) HAPPY, 2) JOYFUL were you feeling right before the
phone went off?” (Figure 1.2). NA was assessed by 3 questions, “How 1) MAD, 2) SAD, 3)
STRESSED were you feeling right before the phone went off?” (Figure 1.3). Responses were
“1=Not at all”, “2=A little”, “3=Quite a bit”, “4=Extremely”. Across all available EMA reports
for each participant, scores for PA were averaged to create a person-level mean of PA
(Chronbach’s alpha=.75); and scores for NA were averaged to create a person-level mean of NA
(Chronbach’s alpha=.79).
Ra, Dissertation
82

82
Covariates
The single parents household status (single-parent vs. dual-parent) was included as a
covariate in the analyses.
For Study 3, Aim 6, data from the H&H study, which was described in Study 2, was used to
explore the moderating effects of physical activity on the association between Digital Media
Use (DMU) and ADHD. At wave 3 (baseline), DMU and physical activity were assessed, and
ADHD symptom status was assessed at waves 4-7.
Physical Activity
H&H study includes four physical activity questions adopted from Youth Risk Behavior
Survey (YRBS). Physical activity was assessed by asking a question, “How many days were you
physically active for at least 60 minutes?” Responses ranged from “None” to “7 days”. This
question was based on the recommendations from the U.S. Department of Health and Human
Services in 2008 that young people aged 6–17 years participate in at least 60 minutes of physical
activity (U.S. Department of Health and Human Services. 2008). Based on the recommendation,
the responses were dichotomized to 7 days (Meet the 60-minute per day MVPA recommendation)
vs. fewer than 7 days (does not meet the 60-minute per day MVPA recommendation).
Past 6-Month ADHD Symptom Status
At each assessment (waves 4-7), students were administered the current ADHD symptom
scale (Barkley, 1991), which contains 18 questions aligned with each of the DSM-IV
(Association, 2013) ADHD symptom criteria: 9 items of inattention symptoms (e.g., difficulty
organizing and completing tasks) and 9 items of hyperactivity/impulsivity symptoms (e.g.,
difficulties remaining still or with task persistence). Respondents rated how frequently they
Ra, Dissertation
83

83
experienced each individual symptom during the past 6 months on a 4-point Likert scale (Never
or Rare, Sometimes, Often, Very Often). Consistent with recommendations (Barkley, 1991) and
DSM-IV criteria (Association, 2013), adolescents who reported experiencing “Often” or “Very
Often” for >6 inattention symptoms or >6 hyperactivity/impulsivity symptoms were classified as
meeting symptom criteria for ADHD within the prior 6-month interval.
Modern Digital Media Use
At baseline, participants were administered a checklist instructing students to select one
of four options to indicate their typical frequency of engaging in 14 digital media activities in the
past week (0 times [none]; 1-2 times a week [low]; 1-2 times per day [moderate]; many times per
day [high]); the activities are listed in Figure 3.3. Responses were categorized into binary options
to distinguish high frequency rate (many times per day) vs. all other responses. The total number
of digital media activities engaged at a high frequency rate were summed to cumulative digital
media use frequency index (range: 0-14), which was the primary exposure variable.
Positive Urgency and Negative Urgency
The UPPS-P Impulsive Scale (Lynam et al., 2007) was used to measure five facets of
impulsivity including positive urgency and negative urgency. The 59-item scale included 12
negative urgency items (e.g., I often act without thinking when I am upset) and the 14-item
positive urgency scale (e.g., I tend to act without thinking when I am really excited) developed
by Cyders and Smith (Cyders & Smith, 2007). The items were randomly ordered, and each item
was rated on a four-point Likert scale ranging from 1 (strongly agree) to 4 (strongly disagree).
The Cronbach’s alphas of the two scales were .96 for PU and .92 for NU (See appendix II for
more details).

Ra, Dissertation
84

84
Covariates
To examine whether associations between digital media use and Symptom status were
independent of potential confounding influences, baseline factors associated with digital media
use or ADHD were selected as covariates.(Connor et al., 2003; Sibley et al., 2011; Yen et al.,
2007) Sociodemographic covariates including age, gender, race/ethnicity, and subsidized lunch
eligibility, which is based in part on the student’s family income level relative to the U.S. federal
poverty guideline (no free lunch [income >185% poverty line], free or reduced lunch [income
<130% - 185% above poverty line]), were assessed with investigator-defined forced-choice items
(see response categories in Table 4.3) (Ra et al, 2018).
Given frequent behavioral and emotional problems among youth with ADHD,(Elia et al.,
2008) additional covariates included baseline delinquent behavior, measured with a sum of
frequency ratings for engaging in 11 different behaviors and conduct disorder problems within
the past 6 months (e.g. “ ste al ing” and “skipping school”; range 1 [never] to 6 [10 or more times];
Cronbach’s alpha internal consistency estimate = .73) (Thompson et al., 2007), past-week
depressive symptoms measured using the Center for Epidemiological Studies Depression Scale
(CESD) (Radloff, 1991) which provides sum score for 20 symptoms (0=rarely or none of the
time to 3=most or all of the time; alpha = .82), self-reported substance use operationalized as a
three-level variable (current [past 30 days] vs. past [ever use, but no use in the past 30 days] vs.
never use of cigarettes, alcohol, or cannabis), (Leventhal et al., 2015) and family history of
substance use (i.e., having siblings, parents, grandparents who smoked cigarettes or had an
alcohol or drug abuse problem; yes/no) (Ra et al., 2018).

Ra, Dissertation
85

85
Statistical Analysis
For Study Aim 5, the prompt-level dataset providing PA and NA at the prompt-level and
the person-level dataset including the person-level ADHD outcome and covariates were merged.
This study employed a two-stage data analysis approach (MIXREGLS-MIXOR models (Hedeker
& Gibbons, 1996, 2006) in which 1) the first stage used to obtain the person-level average and
the variability of PA and NA at the prompt level, and 2) the second stage integrated the effects
from the first stage model into a single-level logistic regression (see Hedeker & Nordgren, 2013
for more details on this modeling approach). The first stage was tested as an empty model with
no covariates. To examine the moderating effects of MVPA, the second stage was tested with the
interaction term between the person-level average and physical activity. The second stage model
controlled for single-parent household status. Two interaction models of PA and NA were
conducted, to test the unique effects of PA and NA on ADHD outcomes.
Main effect model:
Level 1: PA (or NA)_ij = b0_i+ + e_ij
Level 2: ADHDi = ß 0+ ß 1*Single parent household+ ß
2
*Mean
i
+ ß
3
* MVPA
i

Interaction model:
Level 1: PA (or NA)_ij = b0_i+ + e_ij
Level 2: ADHDi = ß 0+ ß 1*Single parent household+ ß
2
*Mean
i
+ ß
3
*MVPA
p

+ ß
4
* Mean
i
_MVPA
i

• i: Person-level, j: prompt-level
Ra, Dissertation
86

86
For Study Aim 6, descriptive analyses of the prevalence and correlation of main study
variables – MVPA, baseline ADHD symptom status, and modern digital media use was first
reported. First, the interaction term of DMU  MVPA was created to test whether the
association between DMU and ADHD symptom status was changed by the level of MVPA.
Given the potential for reciprocal pathways (or selection effects) in which youth with ADHD
may be more liable to seek out digital media usage (Nikkelen et al., 2014), only adolescents with
negative baseline ADHD status (having no ADHD) were included in the analyses. The analyses
used generalized estimating equations (GEEs) (Zeger et al., 1988), utilizing repeated measures
binary (link logit) logistic regressions for dichotomous outcomes in which each participant had
up to four time points of follow-up data (at 6-, 12-, 18-, and 24-months) modeled using an
exchangeable correlation structure (Hubbard et al., 2010). GEEs model the odds of positive (vs.
negative) ADHD symptom status at the follow-ups based on baseline digital media use regressor
values. The set of models used the digital media activities composite score as the primary
regressor. Models were tested with time (6-, 12-, 18-, and 24-month follow-up) and school as
fixed effects and then retested after they are adjusted for baseline covariates (sex, age, ethnicity,
subsidized lunch status, substance use, delinquent behavior and CESD for depressive symptoms).
The digital media use regressor estimate indicates the averaged association with ADHD
collapsed across the four follow-ups. Digital media use (DMU)  Time interaction terms were
added in subsequent models to test whether the association is changed across follow-up time
points.
Since only participants with complete digital media use and ADHD data were included in
the analyses, missing data for the adjustment covariates is permitted (see Table 4.3 for available
data for each covariate) and accounted by multiple-imputation (Rubin, 2004), which replaces
Ra, Dissertation
87

87
each missing value with a set of plausible values that represent the uncertainty about the correct
value to impute. Using the Markov-chain Monte Carlo method for missing at random
assumptions and the available covariate data, five multiple-imputed data sets were created. The
parameter estimates from the models tested in each imputed data set were pooled into a single
estimate. Continuous variables were standardized (M=0, SD=1) to facilitate interpretation of
parameter estimates across covariates with different scaling. Statistical analyses were conducted
using SPSS version 24 (IBM Corp., Armonk, NY). Results were reported as odds ratios (ORs)
with 95% Confidence intervals (CIs). Significance was set to 0.05 and tests were 2-tailed.
RESULTS
Descriptive Statistics for Aim 5
Table 4.1 shows the sample characteristics of baseline and analytic samples of the
MATCH study (N=202 vs. N=168). Final analytic sample size was obtained using listwise
deletion of main variables (affect, ADHD and MVPA).

Ra, Dissertation
88

88
Table 4.1 Sociodemographic characteristics of samples

Baseline Sample
a
Analytic Sample
b
Variable N (%) N (%)
Child Sex
Male 99 (49.0) 90 (53.6)
Female 103 (51.0) 78 (46.4)
Child Age, M (SD) 9.61 (0.91) 9.63 (0.97)
Child Ethnicity
Hispanic 109 (54.0) 92 (54.8)
Non-Hispanic 93 (46.0) 76 (45.2)
Annual Household Income
Less than $35,000 55 (27.2) 44 (26.2)
$35,001-$75,000 59 (29.2) 50 (29.8)
$75,001-$105,000 39 (19.3) 35 (20.8)
$105,001 and above 48 (23.8) 39 (23.2)
MVPA minutes per valid days,
Mean (SD)
59.24 (24.74) 59.48 (25.07)
Positive Affect
c

Mean (SD) 3.02 (0.68) 3.04 (0.67)
Negative Affect
d

Mean (SD) 1.27 (0.27) 1.26 (0.26)
ADHD symptom status
Below borderline
clinical range
175 (86.6) 158 (94)
Within or above
borderline clinical range
11 (5.5) 10 (6.0)
Total N
202 168
Note.
a
The number of observations does not add up to total due to missing data on that variable.
b
Final analytic sample using listwise deletion of main variables.
c
Score range: 1 (Not at all) to 4
(Extremely) of positive affect (the average of feeling happy and joyful) when prompted.
d
Score
range: 1 (Not at all) to 4 (Extremely) of negative affect (the average of feeling mad, sad and
stressed) when prompted. Abbreviations: SD = Standard deviation; N = Sample size.

There were slightly more girls than boys (53.6% vs. 46.4%) in the analytic sample, and
the mean age of participants was 9.63 years old. Over a half of the sample was Hispanic (54.8%).
The measure of physical activity, MVPA minutes per valid days, was 59.48 minutes per day
(SD=25.07) on average in the analytic sample, which is quite close to the daily recommended
Ra, Dissertation
89

89
level of physical activity (60 minutes per day) for youth. The mean of positive affect was 3.04
(3=quite a bit) out of 4, and the mean of negative affect was 1.26 (1=a little) out of 4. The
number of participants who was within or above borderline clinical range was 10 (6%) in the
analytic sample.
Table 4.2. Results from the models of moderating effects of MVPA in the association between
affective states and ADHD symptom status
Positive Affect Negative Affect
Model Variable Estimate
Standard
Error
p-
value Estimate
Standard
Error
p-
value
Unadjusted Intercept -4.20 1.18 <.001 -3.98 0.88 <.001
Mean of affect -1.62 1.22 0.19 -0.96 1.04 0.36
MVPA 0.02 0.02 0.27 0.02 0.01 0.15

Mean X MVPA
Interaction 0.02 0.02 0.18 0.01 0.01 0.23

Adjusted Intercept -4.03 0.87 <.001 -3.93 0.89 <.001

Single parent
household 2.51 1.08 0.02 2.40 0.96 0.01
Mean of affect -0.12 0.38 0.75 -0.15 0.41 0.71
MVPA 0.02 0.02 0.26 0.02 0.01 0.19

Mean X MVPA
Interaction 0.02 0.02 0.18 0.02 0.01 0.22

The associations between positive affect and ADHD symptom status were not
statistically significant in both PA and NA before and after adjusting for single parent household
(p=.75, p-.71 respectively in the adjusted model). There were no moderating effects of MVPA in
both PA and NA (p=.18, p=.22 respectively in the adjusted model). Single parent household was
positively associated with ADHD symptom criteria and was statistically significant in both PA
and NA models controlling for single parent household (β=2.51, p=.02; β=2.40, p=.01
respectively).

Ra, Dissertation
90

90
Table 4.3. Sample Characteristics at Baseline by Baseline ADHD Status

Note.
a
The denominator is for the data in column 2 and is provided due to missing data for this variable in each
column in this row (or rows; the denominators for column 3 is not provided).
b
Score range: 1 (never) to 6 (10 or
more) within the past 6 months for each item x 11 items.
c
Score range: 0 (rarely or none of the time; 0-1 day) to 3
(most or all of the time; 5-7 days) for each symptom x 20 symptoms.
d
Score range: 0 (none) to 3 (many times a day)
for each item x 14 items (example item: “checking social media sites”).
e
Score range: 0 (none) to 3 (very often) for
each symptom x two 9-item subscales (inattentive and hyperactive/impulsive subtypes).
f
Calculated using the χ
2
test.
g
Calculated using the independent samples t test. Abbreviations : CESD, Center for Epidemiologic Studies
Depression Scale, N: Sample size, SD: Standard Deviation.

(N=2,508)
Ra, Dissertation
91

91
Descriptive Statistics for Aim 6
Table 4.3 reports descriptive statistics for recommended physical activity and other
baseline characteristics for the analytic sample of participants with negative baseline ADHD
status (N=2,508) and reports comparisons to students positive baseline ADHD status (N=186).
The two groups did not significantly differ by gender (p=.11), age (p=.17), ethnicity (p=.86), or
subsidized lunch eligibility (p=.69). Students with a positive (vs. negative) ADHD status at
baseline included a greater proportion of ever substance use (p=.01); reported more delinquent
behavior (p <.001); and higher depressive symptoms (p <.001); more frequent digital media use
(p<.001).

Ra, Dissertation
92

92
Table 4.4. Moderating effects of MVPA in the association of baseline digital media use with
ADHD status across 6- 12- 18- and 24-month follow-ups among ADHD-negative
adolescents

Positive (vs. Negative) ADHD Status at
Follow-up
Baseline Regressors OR (95% CI)
a
P Value
Unadjusted Model
b

Digital media use frequency
c,d
1.43 (1.31, 1.56) <.001
MVPA 0.99 (0.87, 1.12) 0.836
Digital media use frequency × MVPA
e
1.05 (0.94, 1.17) 0.4
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.94, 1.01) 0.413
Digital media use frequency × time
e
0.95 (0.93, 0.98) 0.093
Adjusted Model
f

Digital media use frequency
c,d
1.38 (1.15, 1.65)

<.001
MVPA 1.05 (0.93, 1.18) 0.472
Digital media use frequency × MVPA
e
1.05 (0.94, 1.17) 0.398
Time (6- 12- 18- and 24-month follow-ups) 0.97 (0.94, 1.01) 0.452
Digital media use frequency × time
e
0.95 (0.90, 1.01) 0.124
Female (vs. male) gender 0.53 (0.42, 0.67) <.001
Age
d
0.96 (0.89, 1.05)

0.41
Race/ethnicity
Hispanic Ref
Asian 0.97 (0.62, 1.52) 0.881
Black/African American 1.33 (0.75, 2.36) 0.324
White 1.01 (0.67, 1.52)

0.958
Other 1.06 (0.75, 1.50)

0.745
Subsidized lunch eligibility status (yes vs. no) 0.77 (0.57, 1.05) 0.101
Substance use
Never Ref
Past 1.16 (0.87, 1.53) 0.311
Current 1.21 (0.89, 1.65) 0.214
Family history of substance use (yes vs. no) 1.09 (0.85, 1.39) 0.498
Delinquent behavior
d
1.22 (1.08, 1.38)

0.002
CESD scale for depressive symptoms
d
1.61 (1.44, 1.80)

<.001
Abbreviations: CESD, Center for Epidemiologic Studies Depression Scale; OR odds ratio, CI = Confidence Interval.
Note.
a
OR(95% CI) estimates and corresponding p-values from respective regressor parameter estimates for association with
follow-up ADHD status from repeated binary logistic generalized estimating equations, including school fixed effects.
b
Model
included digital media use as primary regressor.
c
Cumulative index of binary frequent digital media use (Range = 0 – 14).
d
Rescaled (M=0, SD=1) such that the ORs indicate the change in odds in the outcome associated with an increase in 1 SD unit on
the covariate continuous scale.
e
Interaction term added in subsequent model; OR’s for other regressors or covariates are from the
model excluding the interaction term.
f
Model included digital media use and other covariates as simultaneous regressors.

Ra, Dissertation
93

93
Moderating Effects of MVPA in the Associations Between Baseline Modern Digital Media
Use and ADHD Status at Follow-Ups among ADHD-Negative Adolescents at Baseline
The unadjusted estimate for the association of the digital media use frequency at baseline
with positive ADHD status at follow-up was statistically significant (OR, 1.43 [95% CI, 1.31-
1.56], p<.001). The associations remained after adjustment for baseline sociodemographic
covariates, substance use, depressive symptoms, and delinquent behavior (OR, 1.38 [95% CI,
1.15-1.65], p<.001). However, neither MVPA nor the interaction term of digital media use
frequency and MVPA was statistically significant in both unadjusted and adjusted models (Table
4.4).

Ra, Dissertation
94

94
DISCUSSION
Overall, the hypotheses for this study were not supported. For specific aim 5, we did not
find the moderating effects of MVPA in the association between affective states (both PA and
NA) and ADHD symptom status nor the main effects of MVPA on ADHD symptom status. To
the best of my knowledge, this is the first study exploring whether affective states and ADHD
symptoms are moderated (or attenuated) by MVPA. Even though there is no evidence on the
moderating effects of MVPA in the association between affective states (both PA and NA) and
ADHD symptom status yet, the relationship between physical activity and affective states as well
as ADHD symptoms have been proven in previous research (Azrin, Ehle, & Beaumont, 2006;
Biddle & Asare, 2011; Gawrilow, Stadler, Langguth, Naumann, & Boeck, 2016; Verret, Guay,
Berthiaume, Gardiner, & Bé liveau, 2012). For instance, MVPA has been shown to acutely
increase positive affective states whilst decreasing negative affective states in children
(Genevieve Fridlund Dunton et al., 2014; Genevieve F Dunton et al., 2014; Liao et al., 2015).
On the other hand, a few studies have shown that physical activity involvement
negatively associates with behaviors related to ADHD (Azrin, Vinas, & Ehle, 2007; Harvey et al.,
2009; A. L. Smith et al., 2013). There could be two possible explanations why the results from
this study are not consistent with previous findings. First, The estimates may have lost
significance due to the smaller sample size. Since the study sample was based on a non-clinical
healthy population without any history of ADHD diagnosis, only 10 participants (6%) were
classified as ‘within or above borderline clinical range’ in the study sample. Also, ADHD
children are more likely to be hyperactive, and have difficulty staying still or remaining calm.
Thus, it is possible that participants who were classified as having ADHD symptom status may
Ra, Dissertation
95

95
have moved around more frequently. Those movements were counted as physical activity in
accelerometer and may have biased the results.
Therefore, in future research, longitudinal analyses of the same models should be
conducted using multi-waves. Moreover, other measures of physical activity need to be tested in
the same models to examine if the estimates and significance change.
For specific aim 6, the hypotheses for this study were not supported. The moderating
effects of MVPA in the association between digital media usage and ADHD symptom status nor
the main effects of MVPA on ADHD symptom status. As aforementioned, findings of a
relationship between digital media use and physical activity are mixed. Those studies indicating
that higher rates of digital media use are negatively related to physical activity used more
traditional digital media measures such as playing video games and TV watching (Koezuka et al.,
2006; Lowry et al., 2002; Motl et al., 2006; Vandewater et al., 2004). However, some studies
failed to find the relationship between the two (Hager, 2006; Lanningham-Foster et al., 2006;
Laurson et al., 2008; Roberts & Foehr, 2004; Snoek et al., 2006). A meta-analysis on a
relationship between digital media use and physical activity in children and adolescents
(Marshall et al., 2004) suggested that it can be explained using neither single markers nor
integrated measures of digital media use as modern digital media become more diverse and
ubiquitous. For example, people tend to stream music when they work out. For this particular
digital media activity, the relationship with ADHD may be positive whereas traditional digital
media use such as TV viewing may be negatively associated with ADHD. Therefore, it is
important to investigate differentiated effects of digital media use and physical activity.

Ra, Dissertation
96

96
Limitations
This study has several limitations that should be considered. For Study Aim 5, the current
analysis does not include context as a time-varying moderator. It is possible that accelerometer
counts activity that we do not consider as physical activity (e.g. walking around the supermarket)
or fail to detect activities occurring in non-wear times (e.g. swimming) (Butte, Ekelund &
Westerterp, 2012; Trost & O’Neil, 2014). Therefore, the results from the current study can only
be generalized to the activities that are captured by accelerometer (e.g. not swimming but
including running errands). There is potential for missing data in any given source like EMA
(affective states) and accelerometer (MVPA) to affect the sample size. For Study Aim 6, physical
activity was only assessed by asking students one simple question in a survey, thus there is a
potential for recall bias and social desirability. In addition, summing binary responses whether
students meet the daily MVPA requirement for each day may not equal to the average daily
MVPA across a week. While this self-report measure of physical activity cannot provide the
exact amount of physical activity engaged in throughout the week, it gives an estimate of each
student’s activity level which has shown to be correlated with accelerometer-assessed physical
activity in other studies (Sallis & Saelens, 2000).
Implications
This study is among the first to examine whether physical activity plays a moderating
role in the relationship between both affect and digital media use with ADHD in children and
adolescents. Results from this study have potential implications to inform future interventions,
and provide additional evidence for policy makers and clinicians who seek to recommend that
Ra, Dissertation
97

97
children and adolescents adhere to sufficient amount of physical activity to decrease the impact
of risk factors like affective states and DMU on the severity of ADHD symptoms.
CONCLUSION
Although numerous studies have attempted to intervene in ADHD symptoms, the
majority have tested ADHD diagnosed patients solely using clinical samples and focused on
post-treatments. However, the chronic and impairing course of ADHD throughout the lifespan
continues to take a significant burden on the lives of patients, their families, and society although
they are successfully treated at earlier ages. Thus, it has been argued that earlier interventions
may alter the series of the disorder and thus avoid many of the long-term consequences of
ADHD (Halperin et al., 2012). Our findings can be used to provide preventive (early)
intervention programs for children and adolescents.

Ra, Dissertation
98

98
CHAPTER 5: OVERALL DISCUSSION & CONCLUSIONS
This research conducted in this dissertation contributes to the overall understanding of the
dynamic relationships between affects, digital media use, physical activity, and ADHD in youth,
using real-time data capture techniques (data from the Mother’s and their Children’s Health
(MATCH) Study) and longitudinal study design (data from the Happiness & Health (H&H)
Study). This research addressed questions related to current issues of ADHD that were
understudied in the literature, and provides additional insight on various risk and protective
factors associated with ADHD. The findings from these studies have important implications for
theory, future research methods, and ADHD prevention and intervention programs and policies.
Summary of Findings
Study 1 examined 1) the relationship between the children’s mean level of PA (and NA)
and ADHD symptom status and 2) the moderating effects of the individual variability of PA (and
NA) in the association between the mean affect and ADHD symptom status. The association
between the children’s mean level of PA and ADHD symptom status was negative whilst the
association between the children’s mean level of NA and ADHD symptom status was positive.
However, neither association was statistically significant. Whereas the individual variability of
PA did not moderate the association between the children’s mean level of PA and ADHD
symptom status, the individual variability of NA did moderate the association between the
children’s mean level of NA and ADHD symptom status, which means children with higher
mean level of NA are more likely to have ADHD symptom status when their variability of NA is
also higher than the average. The results of Study 2 demonstrate that frequent use of various
forms of modern digital media was associated with increased odds of meeting symptom criteria
Ra, Dissertation
99

99
for ADHD over a 24-month follow-up of adolescence in 2,508 students who were not positive
for ADHD at baseline. Moreover, both positive and negative urgency were positively associated
with ADHD symptom criteria, however, neither positive nor negative urgency moderated the
impact of digital media use frequency on ADHD symptom occurrence among those adolescents.
For Study 3, the hypotheses for this study were not supported. For specific aim 5, we did not find
the moderating effects of MVPA in the association between affective states (both PA and NA)
and ADHD symptom status nor the main effects of MVPA on ADHD symptom status.
Theoretical Implications
Key findings from these studies partially support theories in examining the role of affect
in the development of ADHD, and behaviors (digital media use and physical activity) that may
increase or decrease the risk of ADHD. This research addressed the impact of affect (both affect
traits (PU and NU)) and affective states (PA and NA) on ADHD occurrence in depth. According
to Martel’s affect dysregulation theory (2006), deficits in affect regulation may put the child on a
trajectory for further dysregulation and exacerbation of ADHD symptoms (Martel, 2009; Martel
& Nigg, 2006). ADHD follows a developmental trajectory characterized by difficulties with
impulse control and attention that arise early in life (Olson, 1996). The theory suggested that
extreme affect like positive affect, negative affect and affect volatility may be related to ADHD
(Martel & Nigg, 2006; Nigg et al., 2002a, 2002b), yet no studies has explored the dynamic
association among those. This dissertation found that there are moderating effects of affect
volatility (variability) of NA in the association between the mean level of negative affect and
ADHD. Additionally, both PU and NU, which reflect extreme positive or negative affect, are
associated with ADHD.
Ra, Dissertation
100

100
Despite evidence that a few genetic factors are associated with ADHD, recently
researchers have paid more attention on behavioral and environmental factors related to ADHD.
For instance, there has been growing public health concern that increasing digital media usage
may adversely impact youth’s neurobehavioral health such as ADHD due to high speed, level of
stimulation, and ubiquitous use throughout the day. Yet, no studies examined its prospective
association with ADHD using an integrated measure of modern digital media activities. Findings
from this study suggest that modern digital media use may be solely associated with ADHD.
Thus, it may be crucial to explore other behavioral and environmental factors related to ADHD
beside genetic factors.
Methodological Implications
Real-time data capture techniques (data from the Mother’s and their Children’s Health
(MATCH) Study) and longitudinal study design (data from the Happiness & Health (H&H)
Study) were used to address different research questions in two populations at risk for ADHD –
children (8-12 year-old) and adolescents (15-18 year-old). Given the complexity of the dynamic
association among affect, digital media use, physical activity, and ADHD, more robust measures
and methods are needed to fully comprehend the factors that contribute to ADHD. EMA
methods allow us to investigate the dynamic association among, the mean level and variability in
affective states and ADHD. Also, validation of the modern digital media activity scale including
its psychometric properties is warranted. Having a dataset including EMA measures of affective
states, an integrated measure of modern digital media usage, both survey and accelerometer data
of physical activity, and the information on the history of clinical ADHD diagnosis in a
longitudinal study will help to better understand the prospective association among them.
Ra, Dissertation
101

101
Political Implications
Findings from the dissertation studies inform the policy development, which in turn,
inform various interventions. Findings from Study 1 indicate that the mean level of negative
affect and ADHD status are positively associated at higher variability of affective states whereas
the mean level of negative affect and ADHD status are negatively associated at lower variability,
meaning that children whose negative affect are high and fluctuate are identified as those at risk
for ADHD. Policy recommendations may include early identification and early intervention for
children whose affective states are high and fluctuate, developing a valid screening tool to
identify those, educating health professionals, and teachers and parents training to engage in
interventions. Early identification and intervention of children at risk for ADHD is essential.
However, current school interventions focus on children with ADHD and their special treatments
(van Stralen, 2016). Even with successful treatment during childhood and adolescence, the
chronic and impairing course of ADHD throughout the lifespan continues to negatively impact
on the lives of not only children with ADHD but also their families, and society. Thus, early
intervention is crucial in addressing an array of ADHD (Halperin et al., 2012). The results from
this study may help to identify at-risk children prior to the onset of serious ADHD
symptomatology, and provide secondary preventive interventions (e.g. parent training for
children at higher risk of ADHD) to reduce (or eliminate) the likelihood of severe consequences
of ADHD in the future. Therefore, understanding how a child’s mean affective and affective
variability relate to ADHD risk can aid in the early identification of ADHD symptomatology,
leading to earlier intervention and improved outcomes.

Ra, Dissertation
102

102
On the other hand, the findings from Study 2 can add new information on digital media
use, and neurobehavioral health of adolescents, which may assist parents, educators, health
professionals, and policy makers to protect ADHD of the adolescent population through careful
examination and monitoring of digital media. If findings indicate that adolescents’ digital media
use (DMU) may be positively associated with ADHD, guidance on the frequency of digital
media use (i.e. not to engage in digital media activity three times or more per day) for
adolescents may be included in the health or education policy. The policy statement may include
parents’ roles and pediatricians. For instance, parents and pediatricians are advised to work
together to develop a media use plan, which considers their children’s developmental stages to
individualize an appropriate balance for media time and consistent rules about media use, to
mentor their children, to set boundaries for accessing content and displaying personal
information, and to implement open family communication about digital media use (COUNCIL,
2016).
Overall Limitations
Although limitations to each specific study have been addressed, there are broader
limitations that are pertinent to all three studies. First of all, this dissertation used two survey data,
and ADHD symptom status was assessed by survey measures in both studies. Responses on the
rating scale may be consistent with the disorder, but are not sufficient to render a diagnosis of
ADHD. Also, this dissertation study was unable to comprehend the full environmental context
that may contribute to the onset of ADHD. Modern digital media changes rapidly, thus the
activities included in the digital media usage may need to be up-to-date since we used data
surveyed between 2014 and 2016. In addition, distinctions between digital media activities on
computers versus portable devices such as Smartphones and tablets both datasets were not
Ra, Dissertation
103

103
distinguished. Lastly, although this dissertation used data from representative samples of
ethnically diverse American youth, the generalizability of the results from this study may be
limited to the present population and may not apply to other youth populations.
OVERALL CONCLUSION
This dissertation sought to address gaps in ADHD literature. This dissertation is the first
study to examine the two different facets of individual affective states (the mean level and
variability in PA and NA) and their dynamic relationship with ADHD. The results from this
study contributes to identify at-risk children prior to the onset of serious ADHD symptomatology,
and provide secondary preventive interventions (e.g. parent training for children at higher risk of
ADHD) to reduce (or eliminate) the likelihood of severe consequences of ADHD in the future.
On the other hand, this dissertation is also one of the first studies that examined the association
between digital media use, urgency (PU and NU), and ADHD. The findings from this study may
add new information on digital media use, affects, and neurobehavioral health of adolescents.
Given the increasing global ubiquity of digital media technologies, careful examination and
monitoring of digital media by parents, educators, health professionals, and policy makers are
crucial to protect the neurobehavioral health of the adolescent population.

Ra, Dissertation
104

104
REFERENCES

Achenbach, T. M., & Edelbrock, C. (1991). Child behavior checklist. Burlington (Vt), 7.
Al‐Yagon, M. (2009). Comorbid LD and ADHD in childhood: socioemotional and behavioural
adjustment and parents’ positive and negative affect. European Journal of Special Needs
Education, 24(4), 371-391. doi:10.1080/08856250903223054
Anestis, M. D., & Joiner, T. E. (2011). Examining the role of emotion in suicidality: Negative
urgency as an amplifier of the relationship between components of the interpersonal–
psychological theory of suicidal behavior and lifetime number of suicide attempts.
Journal of affective disorders, 129(1), 261-269.
Association, D.-A. P. (2013). Diagnostic and statistical manual of mental disorders. Arlington:
American Psychiatric Publishing.
Atladottir, H. O., Gyllenberg, D., Langridge, A., Sandin, S., Hansen, S. N., Leonard, H., . . .
Bourke, J. (2015). The increasing prevalence of reported diagnoses of childhood
psychiatric disorders: a descriptive multinational comparison. European child &
adolescent psychiatry, 24(2), 173-183.
Azrin, N. H., Ehle, C. T., & Beaumont, A. L. (2006). Physical exercise as a reinforcer to promote
calmness of an ADHD child. Behavior modification, 30(5), 564-570.
Azrin, N. H., Vinas, V., & Ehle, C. T. (2007). Physical activity as reinforcement for classroom
calmness of ADHD children: A preliminary study. Child & family behavior therapy,
29(2), 1-8.
Baldwin, J. S., & Dadds, M. R. (2008). Examining alternative explanations of the covariation of
ADHD and anxiety symptoms in children: a community study. Journal of abnormal child
psychology, 36(1), 67-79.
Ra, Dissertation
105

105
Ballard, M., Gray, M., Reilly, J., & Noggle, M. (2009). Correlates of video game screen time
among males: body mass, physical activity, and other media use. Eating behaviors, 10(3),
161-167.
Barkley, R. A. (1991). Attention-deficit hyperactivity disorder: A clinical workbook (Vol. 2):
Medifocus_com Inc.
Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive functions:
constructing a unifying theory of ADHD. Psychological bulletin, 121(1), 65.
Biederman, J., Monuteaux, M., Kendrick, E., Klein, K., & Faraone, S. (2005). The CBCL as a
screen for psychiatric comorbidity in paediatric patients with ADHD. Archives of Disease
in Childhood, 90(10), 1010-1015.
Billieux, J., Gay, P., Rochat, L., & Van der Linden, M. (2010). The role of urgency and its
underlying psychological mechanisms in problematic behaviours. Behaviour Research
and Therapy, 48(11), 1085-1096. doi:https://doi.org/10.1016/j.brat.2010.07.008
Billieux, J., Van der Linden, M., d'Acremont, M., Ceschi, G., & Zermatten, A. (2007). Does
impulsivity relate to perceived dependence on and actual use of the mobile phone?
Applied Cognitive Psychology, 21(4), 527-537.
Billieux, J., Van der Linden, M., & Rochat, L. (2008). The role of impulsivity in actual and
problematic use of the mobile phone. Applied Cognitive Psychology, 22(9), 1195-1210.
Brenhouse, H. C., & Andersen, S. L. (2011). Developmental trajectories during adolescence in
males and females: a cross-species understanding of underlying brain changes.
Neuroscience & Biobehavioral Reviews, 35(8), 1687-1703.
Ra, Dissertation
106

106
Brondolo, E., Brady, N., Thompson, S., Tobin, J. N., Cassells, A., Sweeney, M., . . . Contrada, R.
J. (2008). Perceived racism and negative affect: Analyses of trait and state measures of
affect in a community sample. Journal of Social and Clinical Psychology, 27(2), 150-173.
Brook, J. S., Brook, D. W., Zhang, C., Seltzer, N., & Finch, S. J. (2013). Adolescent ADHD and
adult physical and mental health, work performance, and financial stress. Pediatrics,
131(1), 5-13.
Brown, T. E., & Brown, D. T. E. (2014). Smart but Stuck : Emotions in Teens and Adults with
ADHD. New York, NY, UNITED STATES: John Wiley & Sons, Incorporated.
Bunford, N., Evans, S. W., & Wymbs, F. (2015). ADHD and emotion dysregulation among
children and adolescents. Clinical child and family psychology review, 18(3), 185-217.
Chen, K. Y., & DAVID R BASSETT, J. (2005). The technology of accelerometry-based activity
monitors: current and future. Medicine & Science in Sports & Exercise, 37(11), S490-
S500.
Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E., . . . Kramer,
A. F. (2006). Aerobic exercise training increases brain volume in aging humans. The
Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 61(11),
1166-1170.
Colcombe, S. J., Kramer, A. F., McAuley, E., Erickson, K. I., & Scalf, P. (2004). Neurocognitive
aging and cardiovascular fitness. Journal of Molecular Neuroscience, 24(1), 9-14.
Connor, D. F., Edwards, G., Fletcher, K. E., Baird, J., Barkley, R. A., & Steingard, R. J. (2003).
Correlates of comorbid psychopathology in children with ADHD. Journal of the
American Academy of Child & Adolescent Psychiatry, 42(2), 193-200.
Ra, Dissertation
107

107
Cyders, M. A., & Smith, G. T. (2007). Mood-based rash action and its components: Positive and
negative urgency. Personality and individual differences, 43(4), 839-850.
Cyders, M. A., & Smith, G. T. (2008). Emotion-based dispositions to rash action: positive and
negative urgency. Psychological bulletin, 134(6), 807.
Dalsgaard, S., Østergaard, S. D., Leckman, J. F., Mortensen, P. B., & Pedersen, M. G. (2015).
Mortality in children, adolescents, and adults with attention deficit hyperactivity disorder:
a nationwide cohort study. The Lancet, 385(9983), 2190-2196.
Danielson, M. L., Bitsko, R. H., Ghandour, R. M., Holbrook, J. R., Kogan, M. D., & Blumberg,
S. J. (2018). Prevalence of parent-reported ADHD diagnosis and associated treatment
among US children and adolescents, 2016. Journal of Clinical Child & Adolescent
Psychology, 1-14.
Davis-Becker, K., Peterson, C. M., & Fischer, S. (2014). The relationship of trait negative
urgency and negative affect to disordered eating in men and women. Personality and
individual differences, 56, 9-14.
Dunton, G. F., Huh, J., Leventhal, A., Riggs, N., Spruijt-Metz, D., Pentz, M. A., & Hedeker, D.
(2014). Momentary Assessment of Affect, Physical Feeling States, and Physical Activity
in Children. Health psychology : official journal of the Division of Health Psychology,
American Psychological Association, 33(3), 255-263. doi:10.1037/a0032640
Dunton, G. F., Huh, J., Leventhal, A. M., Riggs, N., Hedeker, D., Spruijt-Metz, D., & Pentz, M.
A. (2014). Momentary assessment of affect, physical feeling states, and physical activity
in children. Health Psychology, 33(3), 255.
Dunton, G. F., Liao, Y., Dzubur, E., Leventhal, A. M., Huh, J., Gruenewald, T., . . . Intille, S.
(2015). Investigating within-day and longitudinal effects of maternal stress on children's
Ra, Dissertation
108

108
physical activity, dietary intake, and body composition: Protocol for the MATCH study.
Contemporary clinical trials, 43, 142-154.
Elia, J., Ambrosini, P., & Berrettini, W. (2008). ADHD characteristics: I. Concurrent co-
morbidity patterns in children & adolescents. Child and adolescent psychiatry and mental
health, 2(1), 15.
Factor, P. I., Reyes, R. A., & Rosen, P. J. (2014). Emotional impulsivity in children with ADHD
associated with comorbid—not ADHD—symptomatology. Journal of Psychopathology
and Behavioral Assessment, 36(4), 530-541.
Fulk, L., Stock, H., Lynn, A., Marshall, J., Wilson, M., & Hand, G. (2004). Chronic physical
exercise reduces anxiety-like behavior in rats. International journal of sports medicine,
25(01), 78-82.
Gapin, J. I., Labban, J. D., & Etnier, J. L. (2011). The effects of physical activity on attention
deficit hyperactivity disorder symptoms: The evidence. Preventive Medicine, 52, S70-
S74.
Gawrilow, C., Stadler, G., Langguth, N., Naumann, A., & Boeck, A. (2016). Physical activity,
affect, and cognition in children with symptoms of ADHD. Journal of attention disorders,
20(2), 151-162.
George, M. J., & Odgers, C. L. (2015). Seven fears and the science of how mobile technologies
may be influencing adolescents in the digital age. Perspectives on psychological science,
10(6), 832-851.
George, M. J., Russell, M. A., Piontak, J. R., & Odgers, C. L. (2017). Concurrent and
Subsequent Associations Between Daily Digital Technology Use and High‐Risk
Adolescents’ Mental Health Symptoms. Child Development.
Ra, Dissertation
109

109
Gestsdottir, S., & Lerner, R. M. (2008). Positive development in adolescence: The development
and role of intentional self-regulation. Human Development, 51(3), 202-224.
Giedd, J. N. (2004). Structural magnetic resonance imaging of the adolescent brain. Annals of the
New York Academy of Sciences, 1021(1), 77-85.
Gruber, J., Johnson, S. L., Oveis, C., & Keltner, D. (2008). Risk for mania and positive
emotional responding: too much of a good thing? Emotion, 8(1), 23.
Hager, R. L. (2006). Television viewing and physical activity in children. Journal of adolescent
health, 39(5), 656-661.
Halperin, J. M., Bé dard, A.-C. V., & Curchack-Lichtin, J. T. (2012). Preventive interventions for
ADHD: a neurodevelopmental perspective. Neurotherapeutics, 9(3), 531-541.
Hamilton, J. L., Burke, T. A., Stange, J. P., Kleiman, E. M., Rubenstein, L. M., Scopelliti, K.
A., . . . Alloy, L. B. (2017). Trait Affect, Emotion Regulation, and the Generation of
Negative and Positive Interpersonal Events. Behavior Therapy, 48(4), 435-447.
doi:https://doi.org/10.1016/j.beth.2017.01.006
Harpin, V. (2017). Management of ADHD in Children and Young People: Wiley.
Harvey, W. J., Reid, G., Bloom, G. A., Staples, K., Grizenko, N., Mbekou, V., . . . Joober, R.
(2009). Physical activity experiences of boys with and without ADHD. Adapted Physical
Activity Quarterly, 26(2), 131-150.
Hedeker, D., & Gibbons, R. D. (1996). MIXREG: a computer program for mixed-effects
regression analysis with autocorrelated errors. Computer Methods and Programs in
Biomedicine, 49(3), 229-252.
Hedeker, D., & Gibbons, R. D. (2006). Longitudinal data analysis (Vol. 451): John Wiley &
Sons.
Ra, Dissertation
110

110
Hegberg, N. J., & Tone, E. B. (2015). Physical activity and stress resilience: Considering those
at-risk for developing mental health problems. Mental Health and Physical Activity, 8, 1-
7.
Holmes, M. E. (2018). Physical Activity, Stress, and Obesity Handbuch Stressregulation und
Sport (pp. 311-323): Springer.
Hubbard, A. E., Ahern, J., Fleischer, N. L., Van der Laan, M., Lippman, S. A., Jewell, N., . . .
Satariano, W. A. (2010). To GEE or not to GEE: comparing population average and
mixed models for estimating the associations between neighborhood risk factors and
health. Epidemiology, 21(4), 467-474.
Kearnes, T. B., & Ruebel, J. B. (2011). Relationship between Negative Emotion and ADHD
among College Males and Females. Journal of Postsecondary Education and Disability,
24(1), 31-42.
Koezuka, N., Koo, M., Allison, K. R., Adlaf, E. M., Dwyer, J. J. M., Faulkner, G., & Goodman,
J. (2006). The Relationship between Sedentary Activities and Physical Inactivity among
Adolescents: Results from the Canadian Community Health Survey. Journal of
adolescent health, 39(4), 515-522. doi:https://doi.org/10.1016/j.jadohealth.2006.02.005
Kramer, A. F., Erickson, K. I., & Colcombe, S. J. (2006). Exercise, cognition, and the aging
brain. Journal of applied physiology, 101(4), 1237-1242.
Kwon, S.-Y. L. J. H. (2001). Impulsivity, social problem-solving abilities, and communication
style of adolescent internet game addicts. The Korean Journal of Clinical Psychology,
20(1), 67-80.
Ra, Dissertation
111

111
Lanningham-Foster, L., Jensen, T. B., Foster, R. C., Redmond, A. B., Walker, B. A., Heinz, D.,
& Levine, J. A. (2006). Energy expenditure of sedentary screen time compared with
active screen time for children. Pediatrics, 118(6), e1831-e1835.
Larsson, J.-O., Larsson, H., & Lichtenstein, P. (2004). Genetic and environmental contributions
to stability and change of ADHD symptoms between 8 and 13 years of age: a
longitudinal twin study. Journal of the American Academy of Child & Adolescent
Psychiatry, 43(10), 1267-1275.
Laurson, K., Eisenmann, J. C., & Moore, S. (2008). Lack of association between television
viewing, soft drinks, physical activity and body mass index in children. Acta Pædiatrica,
97(6), 795-800. doi:10.1111/j.1651-2227.2008.00713.x
LeDoux, J. (1996). Emotional networks and motor control: a fearful view Progress in brain
research (Vol. 107, pp. 437-446): Elsevier.
Lee, H. W., Choi, J.-S., Shin, Y.-C., Lee, J.-Y., Jung, H. Y., & Kwon, J. S. (2012). Impulsivity in
Internet Addiction: A Comparison with Pathological Gambling. Cyberpsychology,
Behavior, and Social Networking, 15(7), 373-377. doi:10.1089/cyber.2012.0063
Lemola, S., Perkinson-Gloor, N., Brand, S., Dewald-Kaufmann, J. F., & Grob, A. (2015).
Adolescents’ electronic media use at night, sleep disturbance, and depressive symptoms
in the smartphone age. Journal of youth and adolescence, 44(2), 405-418.
Lenhart, A., Duggan, M., Perrin, A., Stepler, R., Rainie, H., & Parker, K. (2015). Teens, social
media & technology overview 2015: Pew Research Center [Internet & American Life
Project].
Ra, Dissertation
112

112
Letzring, T. D., & Adamcik, L. A. (2015). Personality traits and affective states: Relationships
with and without affect induction. Personality and individual differences, 75, 114-120.
doi:https://doi.org/10.1016/j.paid.2014.11.011
Leventhal, A. M., Strong, D. R., Kirkpatrick, M. G., Unger, J. B., Sussman, S., Riggs, N. R., . . .
Audrain-McGovern, J. (2015). Association of electronic cigarette use with initiation of
combustible tobacco product smoking in early adolescence. Jama, 314(7), 700-707.
Liao, Y., Shonkoff, E. T., & Dunton, G. F. (2015). The acute relationships between affect,
physical feeling states, and physical activity in daily life: a review of current evidence.
Frontiers in psychology, 6, 1975.
Lowry, R., Wechsler, H., Galuska, D. A., Fulton, J. E., & Kann, L. (2002). Television viewing
and its associations with overweight, sedentary lifestyle, and insufficient consumption of
fruits and vegetables among US high school students: differences by race, ethnicity, and
gender. Journal of school health, 72(10), 413-421.
Lynam, D., Smith, G., Cyders, M., Fischer, S., & Whiteside, S. (2007). The UPPS-P
questionnaire measure of five dispositions to rash action. Unpublished technical report,
Purdue University.
Maher, J. P., Ra, C. K., Leventhal, A. M., Hedeker, D., Huh, J., Chou, C.-P., & Dunton, G. F.
(2018). Mean level of positive affect moderates associations between volatility in positive
affect, mental health, and alcohol consumption among mothers. Journal of abnormal
psychology, 127(7), 639.
Marmorstein, N. R. (2013). Associations between dispositions to rash action and internalizing
and externalizing symptoms in children. Journal of Clinical Child & Adolescent
Psychology, 42(1), 131-138.
Ra, Dissertation
113

113
Marshall, S. J., Biddle, S. J., Gorely, T., Cameron, N., & Murdey, I. (2004). Relationships
between media use, body fatness and physical activity in children and youth: a meta-
analysis. International journal of obesity, 28(10), 1238.
Martel, M. M. (2009). Research review: a new perspective on attention‐deficit/hyperactivity
disorder: emotion dysregulation and trait models. Journal of Child Psychology and
Psychiatry, 50(9), 1042-1051.
Martel, M. M., & Nigg, J. T. (2006). Child ADHD and personality/temperament traits of reactive
and effortful control, resiliency, and emotionality. Journal of Child Psychology and
Psychiatry, 47(11), 1175-1183.
Meerkerk, G. J., van den Eijnden, R. J. J. M., Franken, I. H. A., & Garretsen, H. F. L. (2010). Is
compulsive internet use related to sensitivity to reward and punishment, and impulsivity?
Computers in Human Behavior, 26(4), 729-735.
doi:https://doi.org/10.1016/j.chb.2010.01.009
Moffitt, T. E., Houts, R., Asherson, P., Belsky, D. W., Corcoran, D. L., Hammerle, M., . . .
Polanczyk, G. V. (2015). Is adult ADHD a childhood-onset neurodevelopmental disorder?
Evidence from a four-decade longitudinal cohort study. American Journal of Psychiatry,
172(10), 967-977.
Motl, R. W., McAuley, E., Birnbaum, A. S., & Lytle, L. A. (2006). Naturally occurring changes
in time spent watching television are inversely related to frequency of physical activity
during early adolescence. Journal of Adolescence, 29(1), 19-32.
doi:https://doi.org/10.1016/j.adolescence.2005.01.005
Ra, Dissertation
114

114
Mottram, A. J., & Fleming, M. J. (2009). Extraversion, Impulsivity, and Online Group
Membership as Predictors of Problematic Internet Use. CyberPsychology & Behavior,
12(3), 319-321. doi:10.1089/cpb.2007.0170
Nigg, J. T. (2006). What causes ADHD?: Understanding what goes wrong and why: Guilford
Press.
Nigg, J. T. (2013). Attention-deficit/hyperactivity disorder and adverse health outcomes. Clinical
psychology review, 33(2), 215-228.
Nikkelen, S. W., Valkenburg, P. M., Huizinga, M., & Bushman, B. J. (2014). Media use and
ADHD-related behaviors in children and adolescents: A meta-analysis. Developmental
psychology, 50(9), 2228.
Nikken, P., & de Haan, J. (2015). Guiding young children’s internet use at home: Problems that
parents experience in their parental mediation and the need for parenting support.
Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 9(1).
Nyarko, K. A., Grosse, S. D., Danielson, M. L., Holbrook, J. R., Visser, S. N., & Shapira, S. K.
(2017). Treated prevalence of attention-deficit/hyperactivity disorder increased from
2009 to 2015 among school-aged children and adolescents in the United States. Journal
of child and adolescent psychopharmacology, 27(8), 731-734.
Okado, I., Mueller, C. W., & Nakamura, B. J. (2016). Positive and negative affect in clinic-
referred youth with ADHD. Journal of attention disorders, 20(1), 53-62.
Palmonari, A., Pombeni, M. L., & Kirchler, E. (1989). Peergroups and evolution of the self-
system in adolescence. European Journal of Psychology of Education, 4(1), 3.
Ra, Dissertation
115

115
Park, J. H., Han, D. H., Kim, B.-N., Cheong, J. H., & Lee, Y.-S. (2016). Correlations among
social anxiety, self-esteem, impulsivity, and game genre in patients with problematic
online game playing. Psychiatry investigation, 13(3), 297-304.
Pedersen, S. L., Walther, C. A. P., Harty, S. C., Gnagy, E. M., Pelham, W. E., & Molina, B. S. G.
(2016). The indirect effects of childhood attention deficit hyperactivity disorder on
alcohol problems in adulthood through unique facets of impulsivity. Addiction, 111(9),
1582-1589. doi:10.1111/add.13398
Perou, R., Bitsko, R. H., Blumberg, S. J., Pastor, P., Ghandour, R. M., Gfroerer, J. C., . . .
Schieve, L. A. (2013). Mental health surveillance among children—United States, 2005–
2011. MMWR Surveill Summ, 62(Suppl 2), 1-35.
Polanczyk, G., De Lima, M. S., Horta, B. L., Biederman, J., & Rohde, L. A. (2007). The
worldwide prevalence of ADHD: a systematic review and metaregression analysis.
American Journal of Psychiatry.
Polanczyk, G. V., Willcutt, E. G., Salum, G. A., Kieling, C., & Rohde, L. A. (2014). ADHD
prevalence estimates across three decades: an updated systematic review and meta-
regression analysis. International journal of epidemiology, 43(2), 434-442.
Ra, C. K., Cho, J., Stone, M. D., De La Cerda, J., Goldenson, N. I., Moroney, E., ... & Leventhal,
A. M. (2018). Association of digital media use with subsequent symptoms of attention-
deficit/hyperactivity disorder among adolescents. Jama, 320(3), 255-263.
Radloff, L. S. (1991). The use of the Center for Epidemiologic Studies Depression Scale in
adolescents and young adults. Journal of youth and adolescence, 20(2), 149-166.
Ra, Dissertation
116

116
Rimmele, U., Seiler, R., Marti, B., Wirtz, P. H., Ehlert, U., & Heinrichs, M. (2009). The level of
physical activity affects adrenal and cardiovascular reactivity to psychosocial stress.
Psychoneuroendocrinology, 34(2), 190-198.
Roberts, D. F., & Foehr, U. G. (2004). Kids and media in America: Cambridge University Press.
Romer, D., & Moreno, M. (2017). Digital media and risks for adolescent substance abuse and
problematic gambling. Pediatrics, 140(Supplement 2), S102-S106.
Romo, L., Ré mond, J.-J., Coeffec, A., Kotbagi, G., Plantey, S., Boz, F., & Kern, L. (2015).
Gambling and Attention Deficit Hyperactivity Disorders (ADHD) in a population of
french students. Journal of gambling studies, 31(4), 1261-1272.
Rosen, P. J., Epstein, J. N., & Van Orden, G. (2013). I know it when I quantify it: ecological
momentary assessment and recurrence quantification analysis of emotion dysregulation
in children with ADHD. ADHD Attention Deficit and Hyperactivity Disorders, 5(3), 283-
294.
Rosen, P. J., Walerius, D. M., Fogleman, N. D., & Factor, P. I. (2015). The association of
emotional lability and emotional and behavioral difficulties among children with and
without ADHD. ADHD Attention Deficit and Hyperactivity Disorders, 7(4), 281-294.
doi:10.1007/s12402-015-0175-0
Rubin, D. B. (2004). Multiple imputation for nonresponse in surveys (Vol. 81): John Wiley &
Sons.
Rusell, J. A. (1980). A circumplex model of affect. Journal of personality and social psychology,
39(6), 1161.
Ra, Dissertation
117

117
Sallis, J. F., & Saelens, B. E. (2000). Assessment of physical activity by self-report: status,
limitations, and future directions. Research quarterly for exercise and sport, 71(sup2), 1-
14.
Sherman, L. E., Payton, A. A., Hernandez, L. M., Greenfield, P. M., & Dapretto, M. (2016). The
power of the like in adolescence: Effects of peer influence on neural and behavioral
responses to social media. Psychological science, 27(7), 1027-1035.
Sherry, J. L. (2004). Flow and Media Enjoyment. Communication Theory, 14(4), 328-347.
doi:10.1111/j.1468-2885.2004.tb00318.x
Sibley, M. H., Pelham, W. E., Molina, B. S., Gnagy, E. M., Waschbusch, D. A., Biswas, A., . . .
Karch, K. M. (2011). The delinquency outcomes of boys with ADHD with and without
comorbidity. Journal of abnormal child psychology, 39(1), 21-32.
Smith, A. L., Hoza, B., Linnea, K., McQuade, J. D., Tomb, M., Vaughn, A. J., . . . Hook, H.
(2013). Pilot physical activity intervention reduces severity of ADHD symptoms in
young children. Journal of attention disorders, 17(1), 70-82.
Smith, G. T., & Cyders, M. A. (2007). Integrating affect and impulsivity: The role of positive
and negative urgency in substance use risk. Drug & Alcohol Dependence, 163, S3-S12.
doi:10.1016/j.drugalcdep.2015.08.038
Snoek, H. M., van Strien, T., Janssens, J. M. A. M., & Engels, R. C. M. E. (2006). The Effect of
Television Viewing on Adolescents’ Snacking: Individual Differences Explained by
External, Restrained and Emotional Eating. Journal of adolescent health, 39(3), 448-451.
doi:https://doi.org/10.1016/j.jadohealth.2005.12.020
Ra, Dissertation
118

118
Spartano, N. L., Heffernan, K. S., Dumas, A. K., & Gump, B. B. (2017). Accelerometer-
determined physical activity and the cardiovascular response to mental stress in children.
Journal of science and medicine in sport, 20(1), 60-65.
Stults-Kolehmainen, M. A., & Sinha, R. (2014). The effects of stress on physical activity and
exercise. Sports medicine, 44(1), 81-121.
Swain, R. A., Harris, A. B., Wiener, E. C., Dutka, M. V., Morris, H. D., Theien, B. E., . . .
Greenough, W. T. (2003). Prolonged exercise induces angiogenesis and increases
cerebral blood volume in primary motor cortex of the rat. Neuroscience, 117(4), 1037-
1046.
Szymanski, K., Sapanski, L., & Conway, F. (2011). Trauma and ADHD–association or
diagnostic confusion? A clinical perspective. Journal of Infant, Child, and Adolescent
Psychotherapy, 10(1), 51-59.
Thompson, M. P., Ho, C.-h., & Kingree, J. B. (2007). Prospective associations between
delinquency and suicidal behaviors in a nationally representative sample. Journal of
adolescent health, 40(3), 232-237.
Troiano, R. P., Berrigan, D., Dodd, K. W., Masse, L. C., Tilert, T., & McDowell, M. (2008).
Physical activity in the United States measured by accelerometer. Medicine and science
in sports and exercise, 40(1), 181.
Van Praag, H. (2008). Neurogenesis and exercise: past and future directions. Neuromolecular
medicine, 10(2), 128-140.
van Stralen, J. (2016). Emotional dysregulation in children with attention-deficit/hyperactivity
disorder. Attention Deficit and Hyperactivity Disorders, 8(4), 175-187.
doi:10.1007/s12402-016-0199-0
Ra, Dissertation
119

119
Vandewater, E. A., Shim, M.-s., & Caplovitz, A. G. (2004). Linking obesity and activity level
with children's television and video game use. Journal of Adolescence, 27(1), 71-85.
doi:https://doi.org/10.1016/j.adolescence.2003.10.003
Verret, C., Guay, M.-C., Berthiaume, C., Gardiner, P., & Bé liveau, L. (2012). A physical activity
program improves behavior and cognitive functions in children with ADHD: an
exploratory study. Journal of attention disorders, 16(1), 71-80.
Visser, S. N., Danielson, M. L., Bitsko, R. H., Holbrook, J. R., Kogan, M. D., Ghandour, R.
M., . . . Blumberg, S. J. (2014). Trends in the parent-report of health care provider-
diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003–
2011. Journal of the American Academy of Child & Adolescent Psychiatry, 53(1), 34-46.
e32.
Webb, H. E., Rosalky, D. S., Tangsilsat, S. E., McLeod, K. A., Acevedo, E. O., & Wax, B.
(2013). Aerobic fitness affects cortisol responses to concurrent challenges. Medicine and
science in sports and exercise, 45(2), 379-386.
Westbrook, R. A. (1987). Product/consumption-based affective responses and postpurchase
processes. Journal of marketing research, 258-270.
Whalen, C. K., Henker, B., Ishikawa, S. S., Floro, J. N., Emmerson, N. A., Johnston, J. A., &
Swindle, R. (2009). ADHD and anger contexts: Electronic diary mood reports from
mothers and children. Journal of pediatric psychology, 34(9), 940-953.
Whiteside, S. P., & Lynam, D. R. (2001). The five factor model and impulsivity: Using a
structural model of personality to understand impulsivity. Personality and individual
differences, 30(4), 669-689.
Ra, Dissertation
120

120
Wiklund, J., Yu, W., Tucker, R., & Marino, L. D. (2017). ADHD, impulsivity and
entrepreneurship. Journal of Business Venturing, 32(6), 627-656.
Yen, J.-Y., Ko, C.-H., Yen, C.-F., Wu, H.-Y., & Yang, M.-J. (2007). The comorbid psychiatric
symptoms of Internet addiction: attention deficit and hyperactivity disorder (ADHD),
depression, social phobia, and hostility. Journal of adolescent health, 41(1), 93-98.
Zeger, S. L., Liang, K.-Y., & Albert, P. S. (1988). Models for longitudinal data: a generalized
estimating equation approach. Biometrics, 1049-1060.
Zheng, F., Gao, P., He, M., Li, M., Wang, C., Zeng, Q., . . . Zhang, L. (2014). Association
between mobile phone use and inattention in 7102 Chinese adolescents: a population-
based cross-sectional study. BMC public health, 14(1), 1022.

Ra, Dissertation
121

121
APPENDIX A. CBCL DSM-V ORIENTED SCALES

Ra, Dissertation
122

122
APPENDIX B. UPPS-P POSITIVE AND NEGATIVE URGENCY QUESTIONNAIRE

Abstract (if available)

Abstract Background: The prevalence of ADHD diagnosed in the United States has been constantly growing among youth in the last two decades. ADHD is one of the most common and high-impact neurodevelopmental disorders during childhood and adolescence. Thus, it is crucial that early identification and intervention of ADHD among at-risk populations may alter the trajectory and long-term consequences this disorder. This study examined the relationships among affect, digital media use, physical activity, and ADHD symptoms, using real-time data capture techniques, and longitudinal study design. The three studies addressed (1) the association between affective states (the mean level and variability of positive and negative affect

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Energy drink consumption, substance use and attention-deficit/hyperactivity disorder among adolescents

PDF

Using a structural model of psychopathology to distinguish relations between shared and specific features of psychopathology, smoking, and underlying mechanisms

PDF

The acute and longitudinal associations between sedentary behaviors, affective states, and emotional disorder symptoms among youth

PDF

The acute relationship between affective states and physiological stress response, and the moderating role of moderate-to-vigorous physical activity

PDF

Understanding the dynamic relationships between physical activity and affective states using real-time data capture techniques

PDF

Role transitions, past life events, and their associations with multiple categories of substance use among emerging adults

PDF

Sociocultural stress, coping and substance use among Hispanic/Latino adolescents

PDF

The role of social support in the relationship between adverse childhood experiences and addictive behaviors across adolescence and young adulthood

PDF

Cultural risk and protective factors for tobacco use behaviors and depressive symptoms among American Indian adolescents in California

PDF

The Internet activities, gratifications, and health consequences related to compulsive Internet use

PDF

Effects of flavorings in electronic cigarettes on the use and appeal of e-cigarettes among youth and adults

PDF

Examining tobacco regulation opinions and policy acceptance among key opinion leaders and tobacco retailers in low socioeconomic status African American, Hispanic, and non-Hispanic White communities

PDF

A sociocultural and developmental approach to intimate partner violence among a sample of Hispanic emerging adults

PDF

The role of depression symptoms on social information processing and tobacco use among adolescents

PDF

Post-traumatic growth among high-risk youth: predictors, impact of stressful life events, and relationship with changes in substance use behaviors

PDF

Social self-control and adolescent substance use

PDF

Objectively measured physical activity and related factors in minority youth

PDF

System dynamics of in-home family eating behavior: insights from intensive longitudinal data using Ecological Momentary Assessment and wearable sensors

PDF

A multi-system evaluation of medication for opioid use disorder for people who use opioids

PDF

Smoke-free housing policies and secondhand smoke exposure in low income multiunit housing in Los Angeles County

Asset Metadata

Creator Ra, Chaelin Karen (author)

Core Title Affect, digital media use, physical activity, and ADHD in youth

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Preventive Medicine (Health Behavior Research)

Publication Date 04/28/2019

Defense Date 03/01/2019

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

Tag ADHD,Adolescents,Children,digital media use,negative affect,OAI-PMH Harvest,physical activity,positive affect,Youth

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Sussman, Steven Yale (committee chair), Schott, Erik Max (committee member), Unger, Jennifer (committee member)

Creator Email chaelinra@gmail.com,chaelinra@hotmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-153605

Unique identifier UC11660750

Identifier etd-RaChaelinK-7330.pdf (filename),usctheses-c89-153605 (legacy record id)

Legacy Identifier etd-RaChaelinK-7330.pdf

Dmrecord 153605

Document Type Dissertation

Format application/pdf (imt)

Rights Ra, Chaelin Karen

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