IMPLICIT ABANDONMENT DISTRESS

BY

JOEL G THOMAS

DISSERTATION

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy in Psychology
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2020

Urbana, Illinois

Doctoral Committee:

Professor Howard Berenbaum, Chair

Associate Professor Florin Dolcos, Co-Director of Research
Professor Wendy Heller, Co-Director of Research
Professor Kelly Bost
Assistant Professor Anita Hund
 ABSTRACT
Although the cognitive revolution has brought mental processes (including implicit
processes) back into psychological science, psychodynamic theory continues to be criticized as
adhering to models of psychopathology that cannot lead to testable predictions and knowledge
accumulation (Hoffart & Johnson, 2017). According to these critics, in a dynamic perspective,
observable indicators of distress and the hypothetical internal mechanism (or “latent entity”)
presumed to cause these symptoms are defined in such a way that they cannot be disentangled
from each other. The current research aimed to address this issue by conceptualizing the internal
mechanism as a relational schema. Focusing on a relational schema of abandonment, it presents
an account of a memory mechanism that can be measured independently of its behavioral
manifestations (i.e., symptoms of distress). In Study 1, an implicit memory script measure,
psychophysiological data, and video capture of behavioral response were used to test the causal
relationship between schema activation and symptoms of distress. In Study 2, the actual
components of the schema were investigated through a Stroop task that aimed to elucidate how
self-other representations of abandonment are connected in memory. The implicit memory script
measure was found to be a better predictor of implicit emotional arousal than any explicit
schema measure. By contrast, the explicit schema measures were better predictors of observed
behavioral responses. Both kinds of schema measures were associated with significant
differences in participant reactions to a non-rejection stimulus rather than a rejection stimulus,
and the representational content associated with abandonment distress did not appear to be
organized in terms of discrete self-other meaning elements. Overall, the findings suggest that
situations in which the meaning of a relational event is open to interpretation may be highly
relevant to how schemas work.

ii
 ACKNOWLEDGEMENTS
This work is a product of a long process of self-discovery and learning that has been
facilitated by many wonderful people. It begins with my own experience in therapy and the
therapist who helped me see myself in ways that were unfamiliar and powerful. That process has
profoundly shaped my interest in psychology—including how I see the world and how I wish to
approach the people in it.
I want to thank my committee members in particular for their steadfast support and
intellectual mentorship. Illinois has been a place where I felt my interests and ideas were both
respected and challenged by faculty who pushed me to clarify my thinking every step of the way.
My initial discussions with Wendy regarding the theoretical basis for this work helped me realize
that how we communicate ideas is just as important as what the ideas are. Her incisive feedback
was my first taste of the kind of intellectual rigor that I hope to pursue throughout my career.
Howard has been a devoted primary research advisor whose guidance was the bedrock for
implementing my ideas in the empirical studies that make up this dissertation. His skills, intense
curiosity, and care are qualities I hope to pass on to my own students some day, and I am grateful
to have been educated as an experimentalist under his mentorship. I want to thank Florin for his
invaluable guidance, commitment of resources, and for welcoming me into his lab to complete
this work, and Kelly for being so giving of her time to help me develop competence in
delivering, coding, and training my research assistants on the ASA measure. Finally, Anita has
been a clinical supervisor who has shown me how to take scientific ideas and make them
relevant to human beings. Her capacity to gently guide me in “putting the pieces together” across
empirical, theoretical, and clinical domains is a true gift. I am so glad to have had her as a
therapy mentor who has helped me feel that pursuing this work has value for improving the lives
of others.
The experience of conducting this research in the company and collaboration of the
thirty-five undergraduate research assistants who worked in my lab is the greatest blessing.
Through dialogue with each other that spanned the technical, theoretical, and personal, the daily
process of conducting research became more play than grind. Feeling connected to those who
you are working with transforms the “work” into something more precious. I will always look
back on the learning environment we co-created as a testament to the importance of cultivating
spaces where our whole selves can be brought, shared, and honored.

iii
 Lastly, I would like to thank my graduate student lifelines. Paul Bogdan, your technical
contributions to the final steps of the analysis made my stress levels swiftly subside. Paul Sharp,
our friendship, intellectual bond, and honesty with each other have been crucial to my personal
and professional development. Rachel Leipow, your love and support has been tremendous in
giving me the strength to complete this project, and it has been a joy to do so with the
companionship of a fellow traveler.

iv
To my undergraduates: Your commitment, curiosity, and contributions
to this work made each day not only possible, but also truly meaningful.

v
 TABLE OF CONTENTS
CHAPTER 1: A NEW MECHANISTIC APPROACH TO RELATIONAL SCHEMAS .............1
CHAPTER 2: EXPERIMENTAL MANIPULATION OF AN ABANDONMENT SCHEMA ....9
CHAPTER 3: IDENTIFYING INFORMATION PROCESSING ELEMENTS OF AN
ABANDONMENT SCHEMA ..................................................................................................... 38
CHAPTER 4: REFLECTIONS ON A NEW MECHANISTIC APPROACH TO
RELATIONAL SCHEMAS ......................................................................................................... 49
REFERENCES ............................................................................................................................. 54
APPENDIX A: IRB LETTERS & DEBRIEFINGS ..................................................................... 59
APPENDIX B: MEASURES........................................................................................................ 63
APPENDIX C: ADDITIONAL TABLES .................................................................................... 83

vi
 CHAPTER 1: A NEW MECHANISTIC APPROACH TO RELATIONAL SCHEMAS
Although early trauma and childhood adversity are well-substantiated antecedents of
adult psychopathology (Van der Kolk, McFarlane, & Weisaeth, 1996), the field of psychology
has yet to converge on a coherent theoretical model of how these negative life experiences are
linked to signs and symptoms of distress in adulthood. For instance, one of the core features of
borderline personality pathology includes “frantic attempts to avoid real or imagined
abandonment.” The memory of early abandonment by a caretaker, now triggered in later life
contexts, represents a possible traumatic etiology for this symptom. A number of other
symptoms of borderline pathology could be considered disturbances in an individual’s
relationship to self and others that could plausibly be explained as consequences of the activation
of early memories of abandonment (e.g., anger outbursts, affect instability, intense interpersonal
relationships, and persistently unstable self-image). Indeed, negative early attachment
experiences have been shown to be associated with a range of adult psychopathology including
borderline personality (Dozier, Stovall-McClough, & Albus, 2008). Nonetheless, no consensus
yet exists regarding the specific mechanisms by which these experiences are carried forward in
time to causally explain later behavior (Fraley, Roisman, & Haltigan, 2013).
The current research aimed to address this issue by conceptualizing the mechanism as a
schema. Chapter 1 begins by describing a few of the central constructs of attachment theory and
shows how a schema approach to linking early relational experiences with signs and symptoms
of distress improves upon the attachment literature in productive ways. Specific methodological
and measurement issues that have hampered the generativity of the schema approach are then
clarified, and a new mechanistic approach that can rigorously test theoretical conjectures from
both the attachment and schema literatures is described. The mechanistic approach described in
Chapter 1 was subsequently implemented in two empirical studies of abandonment schema
activation described in Chapters 2 and 3. General reflections on these studies as initial attempts
to implement a new mechanistic approach to studying relational schemas, as well as their
relevance for psychopathology research and clinical intervention, can be found in Chapter 4.
The attachment approach
Attachment theory was originally formulated to explain the development of emotional
disturbances (Bowlby, 1969). Indeed, early experiences with less sensitive and supportive
caregivers have been demonstrated to lead to less secure attachment (Ainsworth, Blehar, Waters,

1
& Wall, 2015). Furthermore, individual differences in attachment security have been suggested
to be linked with varying symptoms of psychopathology because of associated affect regulation
strategies.
For instance, distressed individuals who are considered securely attached are thought to
expect attachment figures to be available. They seek proximity with such a figure as children,
and as adults may adaptively engage in self-soothing strategies that successfully decrease
distress. Individuals who exhibit an anxious-resistant attachment style are thought to expect that
attachment figures are unavailable, and although seeking proximity, will continue to experience
fear of rejection (i.e., they “hyper-activate” the attachment system leading to even more distress).
In contrast, those with an anxious-avoidant attachment style are thought to avoid proximity when
distressed (i.e., they “de-activate” the attachment system to avoid fear of rejection and other
emotions).
Bowlby (1969) initially explained this affect regulatory influence of attachment over the
lifespan by suggesting that early relational experiences are stored in internal working models. A
“working model” is a self-other representation of caregiving experience that directs future
behavior. Unfortunately, the working model construct has often been used in a metaphoric
manner rather than as a specific statement regarding the representational elements in memory
that may explain individual differences in attachment behaviors (Hinde, 1988; Thompson, 2008).
In particular, it has been argued that current attachment theory fails to explain three key issues.
First, it fails to provide a clear articulation of the content of internal working models. Second, it
fails to explain how such a working model functions to direct behavior in situations that are
stressful or rejecting. Third, the categorical nature of attachment designation fails to explain the
cross-temporal instability of internal working models found in several longitudinal studies
(Pietromonaco & Barrett, 2000; Thompson & Raikes, 2003)
Given these limitations, an increasing number of researchers have begun to assume that
internal working models can be better conceptualized as cognitive schemas. For several reasons,
this kind of formulation is a boon to researchers interested in specifying the mechanisms by
which early life experience can be linked to signs and symptoms of distress in adulthood. First,
because schemas are proposed to contain actual autobiographical elements of early parent-child
experiences, the content of these representations can be articulated in greater detail than in a
working model. Second, this content is thought to be associated with specific information

2
processing functions that can in turn be linked to specific behaviors. This aspect of the schema
construct builds on advances in cognitive science that demonstrate how active schemas influence
attention to, recall, and interpretation of schema-congruent information. Third, cognitive theory
assumes that specific maladaptive cognitive schemas remain latent without influencing a
person’s behavior until a schema-congruent context (e.g., interpersonal rejection) activates that
schema. This “diathesis-stress” characteristic has been argued to be a highly plausible way of
explaining the dynamic, cross-temporal (in)stability of attachment. The following section will
elaborate on the schema construct in order to clarify how researchers have typically
operationalized and measured it.
The schema approach
Broadly speaking, schemas are organized elements of past reactions and experiences that
form a relatively cohesive and persistent body of knowledge capable of guiding subsequent
perception, appraisal, and action (Segal, 1988). Young and colleagues have further specified
early maladaptive schemas as pervasive themes instantiated in information structures composed
of memories, emotions, cognitions, and bodily sensations regarding oneself and one’s
relationship with others (Young, 2003). His research group has identified somewhere between
14-18 schemas (e.g., abandonment, emotional deprivation, emotional inhibition, recognition
seeking, defectiveness) that are understood to form as reality-based representations of a child’s
environment. These schemas are grouped in five schema-domains: Disconnection/Rejection,
Impaired Autonomy/Performance, Impaired Limits, Other-Directedness, and
Overvigilance/Inhibition.
Indeed, previous research has provided evidence that early maladaptive schemas are
associated with a range of psychopathological outcomes. Typically, schemas are assessed using
the Young Schema Questionnaire (YSQ), a self-report inventory that asks individuals to rate the
degree to which particular statements are true descriptions of them (e.g., “I feel like a loner” and
“I’m a failure in many ways”). Endorsement of schema items on the YSQ has been shown to be
predictive of the presence of psychopathology (Rijkeboer, van den Bergh, & van den Bout,
2005). Some schemas have also been shown to be associated with symptoms in theoretically
consistent ways. For instance, schemas involving perceptions of the world as dangerous and
threatening such as mistrust/abuse and vulnerability to harm were associated with paranoia on
the Brief Symptom Inventory (BSI). Those involving apprehension and threat (abandonment,

3
subjugation, failure, and vulnerability to harm schemas) were associated with anxiety, and those
involving hopelessness, inadequacy, and loss (abandonment, failure, defectiveness/shame, and
self-sacrifice schemas) were associated with depression (Calvete, Estévez, López de Arroyabe, &
Ruiz, 2005; Welburn, Coristine, Dagg, Pontefract, & Jordan, 2002).
Recent work has shown that endorsement of abandonment schema items on the YSQ was
positively correlated with personality trait measures of emotional lability, anxiousness,
separation insecurity, depressivity, and suspiciousness (Bach, Lee, Mortensen, & Simonsen,
2015). Moreover, a number of studies have now shown that schemas may be predictive of
attachment status and may even explain a large portion of the variance in adult attachment. For
instance, in a clinical sample, Mason, Platts, and Tyson (2005) found a number of associations
between attachment status as measured by the Experiences in Close Relationships (ECR)
questionnaire and maladaptive schemas as measured by the YSQ. Based on combining the two
insecure attachment dimensions, four categories were created (secure: low scores on both
anxious-resistant and anxious-avoidant scales, preoccupied: high on anxious-resistant only,
dismissing avoidant: high on anxious-avoidant only, and fearful avoidant: high scores on both
anxious-resistant and anxious-avoidant scales). Preoccupied individuals had higher scores on
Disconnection/Rejection, Impaired Autonomy and Performance, and Other-Directedness
domains, while fearfully avoidant individuals had higher scores on all five domains.1 Using
discriminant function analysis, they further found that 77% of their sample could be reliably
categorized into an attachment style according to their schemas.
Finally, in a large sample of late adolescents (N = 298), Bosmans, Braet, and Van
Vleirberghe (2010) examined whether early maladaptive schemas can explain the relation
between attachment anxiety and avoidance dimensions and symptoms of psychopathology.
Using a non-parametric sampling approach, they investigated whether the association between
attachment and psychopathology was mediated by early maladaptive schemas. Their results
indicated that the association between attachment anxiety (as measured by the revised ECR) and
psychopathology (as measured by the Symptom Check List-90) is fully mediated by
endorsement of Disconnection/Rejection and Other-Directedness schema items on the YSQ. In
addition, the associations between attachment avoidance and psychopathology were partly
mediated by endorsement of Disconnection/Rejection schemas.

1
The sample contained too few participants in the dismissing avoidant group to perform the same analyses.

4
 Although these findings are supportive of the general idea that experiences related to
attachment may be encoded in schemas, associations between schema items on the YSQ and
measures of psychopathology do not in themselves tell us much about the causal role of these
schemas in the generation of psychological distress. First, schemas have been theoretically
conceptualized as dimensional and latent (unconscious) structures that influence attention to,
recall, and interpretation of schema-congruent information. The YSQ, however, is a self-report
measure that indexes an individual’s observation of their own behavioral patterns. Many argue
that schemas should therefore be measured through projective tests and physiological indicators
of information processing rather than self-report (Muran, 1991; Muran, Samstag, Segal, &
Winston, 1998; Segal & Shaw, 1986; Welburn et al., 2002). In other words, although individuals
likely have some explicit knowledge of their own behavioral patterns (e.g., a person may be able
to observe that they become “clingy” when relationships end), the YSQ appears to tap this
knowledge rather than the actual cognitive functions that are driving their behavior.
The consequence is that the YSQ is likely a “fuzzy” and perhaps distorted measure of the
hypothesized memory structures it is presumably designed to tap. For instance, in Bosmans et
al.’s (2010) study, the “abandonment schema” did not in itself mediate a unique association
between attachment and symptoms (i.e., the relationship between attachment anxiety and
psychopathology was fully mediated by all five Disconnection/Rejection schemas: abandonment,
mistrust/abuse, emotional deprivation, social isolation, and defectiveness subscales of the YSQ).
This issue is particularly pronounced when investigating more severe manifestations of
personality pathology. Studies of those diagnosed with borderline personality disorder, for
instance, have found that these individuals often endorse enough items on the questionnaire that
they meet the threshold for possessing almost all 14-18 schemas on the YSQ.
This lack of specificity in symptom-schema associations may also be due to the inherent
limitations of cross-sectional designs. In such studies, the dynamic nature of the relationship
between symptoms and schemas is left unexplored. As stated earlier, if an abandonment schema
is a latent memory structure, then the activation of the schema is likely to occur most visibly in
moments of perceived interpersonal rejection. Given these problems in methodology and
measurement, researchers have consequently argued that a useful next step would be to
investigate the specific ways in which attachment experiences are instantiated in memory such
that they shape moment-to-moment behavior (Fraley et al., 2013). Such an account of schemas

5
requires an integrative approach that bridges psychophysiology with the cognitive, affective, and
behavioral components of representational processes.
A new mechanistic approach (schema 2.0)
In line with this objective, researchers have recently investigated how early caregiving
experiences are instantiated in memory through attachment related scripts (Steele et al., 2014). A
script is considered a memory structure that summarizes commonalities across multiple
experiences of certain events. For example, a “restaurant script” involves a series of events that
unfold when entering a restaurant (e.g., look at the menu, order food, eat, pay, leave). This script
then comes to guide expectations and behavior in future visits (Nelson, 1999; Schank & Abelson,
2013). Waters and colleagues proposed that an individual’s history of care and attachment
support is similarly represented in memory as a secure base script (Waters & Waters, 2006). To
assess individual differences in access to this secure base script, they developed the Attachment
Script Assessment (ASA).
The ASA uses word-prompt outlines in which participants tell stories with attachment-
related themes from a set of words. Secure base script “knowledge” is then defined as the degree
to which these narratives reflect an encounter with an attachment-related crisis, a clear
communication of need for assistance, the provision of competent help (e.g., support and
emotional soothing), and a return to normalcy with resolution of the problem (Steele et al., 2014;
Waters & Waters, 2006). The ASA has demonstrated adequate test-retest reliability (Vaughn et
al., 2006) and has shown theoretically meaningful associations with implicit and explicit
measures of adult attachment (Coppola, Vaughn, Cassibba, & Costantini, 2006; Dykas,
Woodhouse, Cassidy, & Waters, 2006; Mikulincer, Shaver, Sapir-Lavid, & Avihou-Kanza,
2009). In addition, longitudinal work indicates that mother-child attachment in the first three
years of life and caregiver sensitivity across childhood and adolescence is significantly
associated with the development of secure base script knowledge (Steele et al., 2014).
Interestingly, McLean, Bailey, and Lumley (2014) examined relationships between the
ASA and the YSQ in a sample of undergraduates (N = 146) and found that secure base script
scores were negatively associated with three of the five Disconnection/Rejection schemas
(emotional deprivation, social isolation, and defectiveness), but showed no association with two
others (mistrust/abuse and abandonment). The lack of significant associations with the latter two
schemas is curious given the fact that the content assessed by these items on the YSQ would

6
likely relate to levels of secure base script knowledge. In their discussion of these results, Mclean
et al. suggest that their modified methodology of eliciting secure base script knowledge through
coding of participant descriptions of three current relationships (rather than responses to the
standard word prompt outlines) may have influenced their findings. Overall, their results
highlight the equivocal relationship between explicit and implicit measures of abandonment.
Although much work has been done to validate the ASA as a measure of the presence of
secure base script, relatively little work has been done to examine the additional representational
elements in memory that may causally explain the behavior of individuals who lack a
consolidated script. The two empirical studies described below begin this work by proposing a
mechanistic model of an abandonment schema that contains testable hypotheses regarding the
representational elements that can account for the operation of the schema in particular contexts.
Figure 1 below illustrates the basic approach.

Figure 1. A functional decomposition of an individual’s self-structure into various schemas, as well as

proposed components of an abandonment schema.

An individual’s self-structure can be characterized as the series of representations of self that a

person has stored in memory. This self-structure can be further decomposed into various parts,
some of which are particularly relevant for explaining the effects of early abandonment on

7
functioning as an adult. From a theoretical perspective, the abandonment schema can be
functionally decomposed into at least three components. First, due to a lack of emotional support
and care early on in life, the individual has encoded an implicit expectation that care and
attention will be withheld or withdrawn in close relationships. Second, an implicit memory of
abandonment distress is represented in memory and is activated when the individual experiences
rejection later on in life. This is a phenomenal experience associated with the early memory of
hopelessness, isolation, despair, and depression. Finally, the individual will lack a consolidated
secure base event-script.
The proposed components of an abandonment schema described above were investigated
in two empirical studies. Study 1 examined the dynamic role of an abandonment schema in
directing experience and behavior following rejection. Study 2 focused on identifying the
specific kind of representational content that may be involved in abandonment schema
activation.

8
CHAPTER 2: EXPERIMENTAL MANIPULATION OF AN ABANDONMENT SCHEMA
Introduction
Study 1 used a social rejection paradigm to test a specific counterfactual dependency
contained in Figure 1 (Chapter 1): if an individual exhibits a less consolidated secure base script,
then they will be more likely to display pronounced signs of distress. In addition, pronounced
distress should be observed only when the individual’s expectation of abandonment is triggered
by a stimulus that is perceived as a social rejection. In this study, participants were introduced to
a stranger that matched their preferred romantic partner gender, a confederate, with whom they
were going to converse during two short sessions (of 10 and 5 min respectively). Following a
pleasant initial interaction, the participant was informed that the confederate did not want to
continue with the experiment. No explanation was given for the confederate’s decision. We
expected that individuals low on secure base script would be more likely to report feelings of
rejection and display signs and symptoms of distress. Half of the sample was exposed to this
experimental rejection condition, and the other half was exposed to a control non-rejection
condition in which they were told that the interaction had to end early because of time
constraints. This condition was not expected to induce significant feelings of distress in either
high or low secure base script individuals.
Our first group of predictions focused specifically on understanding the intensity of
distress exhibited by participants across each condition following the manipulation. Given that
schemas have been conceptualized as unconscious (i.e., implicit) memory mechanisms, we
predicted that the intensity of distress would be better predicted by the ASA (an implicit schema
measure) than any YSQ schema (explicit schema measure) both in terms of blind rating of
observed distress and skin conductance response. Since individuals may be partially aware of
some of their interpersonal patterns (e.g., becoming “clingy” when relationships end), we further
predicted that observed distress might be associated with some of the YSQ schemas. In order to
remain consistent with the findings of McLean et al. on the relationship between ASA scores and
Disconnection/Rejection schemas as measured by the YSQ, we predicted that the emotional
deprivation, social isolation, and defectiveness schemas would be associated with observed
distress while the mistrust/abuse and abandonment schemas would not.
Our second group of predictions focused specifically on the phenomenology of response.
In particular, we were interested in the predictive validity of the ASA in relation to changes in

9
expectations of others, anger, and self-esteem following the experimental manipulation. Based
on theoretical understandings of attachment, we conjectured that individuals who exhibit an
anxious-avoidant interpersonal style (i.e., those who tend to de-activate the attachment system
when experiencing threat) would react with increased anger in order to externalize feelings of
rejection. We further conjectured that individuals who exhibit an anxious-resistant interpersonal
style (i.e., those who tend to hyper-activate the attachment system when experiencing threat)
would react with decreased self-esteem due to preoccupation with the personal significance of
rejection. Finally, we conjectured that participants would exhibit negative expectations of others
following rejection regardless of attachment style. This prediction is consistent with the
theoretical contention that adult attachment designations capture how individuals attempt to self-
soothe following rejection rather than how their perception of others changes.
To summarize, our first set of hypotheses were aimed at clarifying whether secure base
script predicts the intensity of distress associated with social rejection better than self-report
schemas:
• Hypothesis 1: We predicted that the emotional deprivation, social isolation, and
defectiveness subscales of YSQ would be associated with observed distress, but the
abandonment and mistrust abuse subscales would not.
• Hypothesis 2: We predicted that observed facial/non-verbal distress would be better
predicted by low secure base script (i.e., the ASA measure) than endorsement of any of
the five self-report Disconnection/Rejection schema subscales of the YSQ.
• Hypothesis 3: We predicted that skin conductance response would be better predicted by
low secure base script (i.e., the ASA measure) than endorsement of any of the five self-
report Disconnection/Rejection schema subscales of the YSQ.
Our second set of hypotheses addressed whether secure base script knowledge would predict
shifts in phenomenological perspective following social rejection in attachment-specific ways:
• Hypothesis 4: We predicted that individuals low on secure base script knowledge would
be more likely to express negative affective expectations of others following the
manipulation in the rejection condition compared to the non-rejection regardless of self-
reported relationship style.
• Hypothesis 5: We predicted that only individuals low on secure base script knowledge
and those who endorsed an anxious-avoidant (i.e., dismissing) relationship style would

10
 display anger following the manipulation in the rejection condition compared to the non-
rejection.
• Hypothesis 6: We predicted that only individuals low on secure base script knowledge
and those who endorsed an anxious-resistant (i.e., preoccupied) relationship style would
show a decrease in self-esteem following rejection.
Method
Participants
A power analysis was conducted to determine the necessary sample size to detect a
medium effect of the rejection manipulation compared to control (Cohen’s f = 0.25, α = 0.05,
power = 0.80; total sample size = 128). A sample of 100 per condition was chosen based on this
analysis and feasibility. The 200 participants for the study were subsequently recruited through
the subject pool at the University of Illinois at Urbana-Champaign (UIUC). Inclusion criteria
included agreement through written informed consent to participate in a two-part study and
documentation that the participant was over the age of 18. There were no explicit exclusion
criteria for the study. Participants were told that the study aimed to investigate biological
markers and social factors that influence the way people construct narratives, and that the
researchers were particularly interested in how people form initial impressions of each other.
Participants were then randomly assigned to one of two conditions: the experimental (rejection)
condition and a control (non-rejection) condition. The study was approved by the University of
Illinois at Urbana-Champaign Institutional Review Board, and included a detailed description of
how care would be taken to explain the nature of the deception (see below) at debriefing.
The undergraduate participants ranged from age 18-22 (one person was age 39) and 69%
identified as female (31% male). In terms of ethnic background, 19% identified as Hispanic or
Latinx, 79% identified as Not Hispanic or Latinx, and 2% Unknown or Prefer Not to Report. In
terms of racial background, 54% identified as White, 28% Asian, 7% Black or African-
American, 6% Indian Subcontinent, 4% Mixed Race, 3% Pacific Islander, 1% Native American,
and 5% other. 64% described their relationship status as single, 18% casually dating, 19% in a
long-term relationship, and 1% married. We assessed participants’ preferred romantic partner
gender in order to assign a confederate for Part II that matched this preference, creating a higher
likelihood that participants would develop positive expectations for a second interaction. 30%

11
identified their preferred romantic partner gender as female, 67% as male, and 3% as other (for
these individuals we alternated assigning a female or male confederate).
Experimental Procedure
In Part I of the study, participants came into the psychology building for a 20-minute
session during which a research assistant (RA) led them through informed consent. The consent
form stated that the first session would include an online demographic form, an audio-recorded
exercise where they would be asked to form stories from sets of word prompts, and four online
surveys in which the participant would be asked about the degree to which certain statements
describe them. Participants were also informed that they would come to another institute on
campus within one week for Part II—a social interaction component of the study that would take
another 60 minutes. In this session, they would engage in a 10-minute and 5-minute interaction
during which they would get to know another participant and would be connected to a skin
conductance (SC) measurement device for the second (5-minute) interaction. If the participant
consented to the study, the RA guided them through the completion of six measures: an online
demographic form, Attachment Script Assessment (ASA), trait Positive and Negative Affect
Scale (T PANAS), Big-Five Personality Questionnaire (PQ), Relationships Questionnaire (RQ),
and abbreviated Young Schema Questionnaire (YSQ).
For Part II of the study, the participant arrived at the Beckman Institute and was brought
by an RA to the assigned study room with a table and two chairs as well as audio and video
recording equipment. The RA reminded the participant that since the study was about how
people form impressions of each other, they would have two brief sessions during which they
would “get to know another participant.” The lab RA then told the participant that they would be
completing the same tasks that the other participant was completing in the room next door prior
to engaging in the interaction. Two electrodes were connected to the participant’s fingers and a
baseline Skin Conductance (SC) measurement was taken for five minutes. After this was
completed, the participant was disconnected from the SC device and the RA had the participant
complete an online state PANAS (S PANAS) questionnaire and a Social Cognition and Object
Relations Scale (SCORS) task that involved telling audiotaped stories to six Thematic
Apperception Test (TAT) cards.
The RA then left the room to get the confederate (another lab assistant whose gender
matched the participant’s romantic partner interest as identified in demographic form) from the

12
room next door. The confederate’s fingers were also taped with two electrodes and the
confederate was told to act as another participant in the study by conversing in a friendly manner
and letting the actual participant lead the conversation. The RA told the participant and
confederate that they would knock on the door when 10 minutes have elapsed. The RA entered
the room after this period of time had elapsed, thanked the participants for completing the first
interaction task, and stated that each of them would now complete a questionnaire on how well
the interaction was going (interaction questionnaire). The RA then asked the confederate to
follow them to complete the questionnaire in the room they were in originally and left the study
room door ajar.
When the RA returned to the room, the participant was connected again to the SC device
and the recording was turned on. The manipulation was then delivered. For every other
participant, the assistant went to get the confederate, and returned in 2 minutes to state, “I’m
sorry but [the confederate] does not want to continue with the second part of the experiment. I’m
not sure what to do right now, so I’ll need to check with my supervisor regarding what to do
next. Please leave your hand still and the equipment connected, and I should be back in a few
minutes.” In the alternating non-rejection condition, the RA stated, “I’m sorry but there’s been a
delay and there is not enough time for the second interaction. I’m not sure what to do right now,
so I’ll need to check with my supervisor regarding what to do next. Please leave your hand still
and the equipment connected, and I should be back in a few minutes.” In both conditions the RA
clicked the escape key on the SC recording laptop in order to record the time at which the
manipulation was delivered so that this information was stored along with the associated SC
data. The RA then left the room and completed an observed reaction measure of the participant’s
facial expression (see observed reaction form).
The supervisor of the study, Joel Thomas (blind to the condition), entered the room 5
minutes later and told the participant that he was the supervisor for the study and that the
participant can still complete the final two tasks. He then disconnected the SC device, left the
room, and completed an observed reaction measure of the participant’s facial expression (see
observed reaction form). The RA then returned and administered again the S PANAS and
SCORS measures. Once these tasks were completed, the supervisor returned to the room to
conduct a funnel debriefing with the participant (i.e., four questions to determine if the
manipulation was believed: “How are you feeling?” “Did you notice anything problematic

13
during the study?” “What did you think the experiment was about?” “What were you thinking
and feeling when the RA told you that you would not be meeting again with the other
participant?”). If answers to any of these questions indicated clear knowledge of the deception,
participant data was not used for subsequent analyses (see manipulation check in results section).
The debriefing form was then presented which described the confederate’s true identity and the
nature of the experimental manipulation. The confederate was brought back in and reintroduced
as the supervisor assured the participant that the confederate had not known what the RA was
going to tell the participant after the interaction. Any remaining concerns were addressed and the
participant was thanked and given subject pool credit for completing the study.
Measures
BACKGROUND INFORMATION
Demographic Form – Part I of the study began with participants completing a demographics
questionnaire that was intended to gather information regarding participants’ age, gender, college
year, time living in the United States, racial/ethnic background, relationship status, preferred
romantic partner gender, English language ability, and whether the participant was interested in
participating in future paid studies.
MANIPULATION CHECK MEASURES
Interaction Questionnaire – This questionnaire was administered after the interaction in Part II
and before the experimental manipulation. It consisted of two questions designed to check that
the interaction was viewed positively enough to reinforce expectations of a second meeting. The
questions were: “Overall, how well do you feel the first interaction period went?” (5-point scale:
1 = very poorly; 5 = very well) and “Are you looking forward to meeting the other person
again?” (Check box: Yes, No, Indifferent). These questions were identical to a previous study on
social rejection that used the same paradigm as the current research proposal (Downey &
Feldman, 1996).
Rejection Items – To assess the degree to which the manipulation in the experimental condition
led participants to feel rejected compared to the control condition, five self-report rejection items
(unaccepted, rejected, hurt, disliked, discouraged) were mixed in to the State PANAS and
administered to the participant before and after the manipulation. Items were rated on a 5-point
scale (1 = Very slightly or not at all; 5 = Extremely). The rejection items were also identical to
those used by Downey and Feldman (1996).

14
State Positive and Negative Affect Scale (S PANAS) – In order to disguise the assessment of
rejection across experimental and control conditions, a 31-item S PANAS was administered
before and after the manipulation. The S PANAS asked participants to indicate to what extent
they feel each emotion at this moment. Previous research suggests that the PANAS has good
convergent and discriminant validity with other measures of psychological functioning and
distress (Watson, Clark, & Tellegen, 1988). The five rejection items (see manipulation check
above) and six low arousal items (three positive: happy, satisfied, tranquil, and three negative:
frustrated, down, sad) were added to the standard 20-item PANAS to create the final measure.
Items were rated on a 5-point scale (1 = Very slightly or not at all; 5 = Extremely). The rejection
items were analyzed to assess the effectiveness of the manipulation, but the other mood
dimensions were not analyzed further in this study.
SCHEMA MEASURES
Abbreviated Young Schema Questionnaire (ABB YSQ) – Five Disconnection/Rejection
schemas were measured using an abbreviated (i.e., 25-item) Young Schema Questionnaire. The
five items for each schema are taken from the YSQ-short form as these items have demonstrated
the highest factor loadings for each schema construct (e.g., abandonment – “I need other people
so much that I worry about losing them”, mistrust/abuse – “I feel that people will take advantage
of me”, emotional deprivation – “I don’t have people to give me warmth, holding and affection”,
social isolation – “I always feel on the outside of groups”, and defectiveness – “No man/woman I
desire could love me once he or she saw my defects”). Items are rated on a 6-point scale (1 =
Completely untrue of me; 6 = Describes me perfectly). The YSQ is a reliable measure that has
demonstrated good internal consistency, good test-retest reliability, and strong construct and
discriminant validity (Hoffart et al., 2005; Rijkeboer et al., 2005; Schmidt, Jr, Young, & Telch,
1995; Van Vlierberghe, Braet, Bosmans, Rosseel, & Bögels, 2010).
Attachment Script Assessment (ASA) – To assess secure base script knowledge, the narrative-
based ASA was administered using four word-prompt cards in which participants told stories
with attachment-related themes from a set of words. Each card has a title for each story and a list
of 12 words (subdivided into three columns). Participants were told to tell the best story they
could tell using the words as an outline and were instructed to speak for approximately 1-2
minutes for each of four cards (two adult-child narratives and two adult-adult narratives). The
ASA has demonstrated strong reliability, validity in terms of consistent associations with

15
attachment interview data, and cross-time stability (Vaughn et al., 2006). Narratives were
transcribed and coded on a 7-point scale (1 = no secure base script with atypical script content; 7
= rich and detailed secure base script) for the extent to which they demonstrated secure base
knowledge by five trained coders (i.e., the lead researcher and four undergraduate assistants).
The lead researcher was trained over two days by a faculty member at the University of Illinois
at Urbana-Champaign (Dr. Kelly Bost, PhD) who previously trained and worked closely with Dr.
Harriet Waters, PhD, one of the co-creators of the instrument and coding manual. The lead
researcher subsequently conducted a two-day training of a team of six undergraduate assistants,
and then selected the four coders who had the most accurate blind ratings of the practice
examples from the ASA scoring manual. The five-member team met weekly during the study to
review coding of participant data and to address any significant discrepancies (i.e., ratings that
differed by more than two points). Inter-reliability for ASA coding was excellent, with an intra-
class correlation coefficient (average measures, absolute agreement) of 0.93 (see Table 1 below).
Table 1. Intra-Class Correlation Coefficients for Coded Variables

Variable ICC
Adult Script Assessment (ASA) 0.93
Video Coding of How Upset 0.61
SCORS – Affective Quality of Representations 0.86
SCORS – Experience and Management of Aggressive Impulses 0.90
SCORS – Self-Esteem 0.68

Note. SCORS = Social Cognition and Object Relations Scale and ICC = Intra-class Correlation Coefficient
(average measures, absolute agreement). ICCs greater than 0.74 are considered excellent; from 0.60 to
0.74, good; from 0.40 to 0.59, fair; and less than 0.40, poor (Fleiss, 1981).

OBSERVATIONAL DISTRESS MEASURES

Observed Reaction Form – To assess observed distress, the research assistant running the
experiment answered the question, “Which of the following is most descriptive of the
participant’s facial expression post-manipulation?” immediately after delivering the
manipulation in each condition. For each participant the research assistant checked one of the
following boxes: 1 = Not upset, 2 = Slightly upset, 3 = Upset, 4 = Very upset. The same form
was completed five minutes after the manipulation by the supervisor and lead researcher for the
study, Joel Thomas, who was blind to the condition. Only the latter ratings were analyzed in this
study.

16
Video Observation Form – To assess observed distress, a 1-minute pre- and post-manipulation
video clip was created for each participant. These clips were randomized and then coded by blind
raters who checked one of the following boxes: 1 = Not upset, 2 = Somewhat upset, 3 = Very
upset. Five coders were originally trained using 12 practice clips that were selected to depict
reactions spanning a range of distress. The ratings of the three coders with the highest degree of
reliability on the actual participant data were used for the analysis. Inter-rater reliability for video
coding was in the good range (ICC = 0.61, see Table 1 above).
IMPLICIT AND PSYCHOPHYSIOLOGICAL DISTRESS MEASURES
Social Cognition and Object Relations Scale – Global Rating Method (SCORS-G) – To assess
change in negative affective expectations of others, anger, and self-esteem, participants tell
stories to six Thematic Apperception Task cards pre-manipulation and another six post-
manipulation. The SCORS has shown adequate reliability, construct validity, and convergent
validity (Hibbard, Hilsenroth, Hibbard, & Nash, 1995; Peters, Hilsenroth, Eudell-Simmons,
Blagys, & Handler, 2006). The lead researcher conducted a nine-week training procedure with
six undergraduate assistants in line with the instructions in the manual for the SCORS-G created
by Stein, Hilsenroth, Slavin-Mulford, and Pinsker (2011), and selected the two coders who had
the most consistent ratings with the practice examples in the scoring manual to score participant
data. The participant stories were randomized so that the three raters (i.e., lead researcher and
two research assistants) were blind to both condition and whether the stories were pre- or post-
manipulation. The response to each card was rated on a 7-point scale (1 = Very disturbed; 7 =
Positive, realistic, and psychologically-minded) on the following subscales of the SCORS-G:
Affective Quality of Representations of Others, Experience and Management of Aggressive
Impulses, and Self-Esteem. The three-member team met weekly during the study to review
coding of participant data and to address any significant discrepancies (i.e., ratings that differed
by more than two points). Inter-rater reliability for the SCORS variables was excellent for
Affective Quality of Representations of Others and Experience and Management of Aggressive
Impulses (ICC = 0.86 and 0.90, respectively) and good for Self-Esteem (ICC = 0.68) (see Table
1 above).
Skin Conductance Response (SCR) – SCR involves applying an electrical potential on two skin
locations and then measuring the current between the two. It is typically understood to represent
level of arousal, as increased activity in the sympathetic nervous system causes increased sweat

17
and thus increased conductance. To control for individual differences in sweat glands and
environmental factors such as humidity, skin conductance responses are often analyzed as
changes in average amplitude of conductivity associated with a stimulus relative to a baseline
conductivity measurement (e.g., Hur, Iordan, Berenbaum, & Dolcos, 2016; Rosebrock, Hoxha,
Norris, Cacioppo, & Gollan, 2017). Then, experimental manipulations are compared in terms of
the changes induced. Researchers also often measure the latency of a skin conductance response
as the time between stimulus onset and the rise of conductivity.
To assess change in physiological arousal in the current study, five minutes of resting
electrodermal activity was recorded at the beginning of part II of the study. Two minutes of
recording was taken just before the manipulation and another five minutes post-manipulation.
SCR was recorded at a sampling rate of 250 Hz using a Biopac MP150 system (Biopac systems
Inc., Goleta, CA, USA), two 11-mm inner diameter Ag/AgCl disposable snap pre-gelled
electrodes, and additional BioPac GEL100 isotonic electrode gel. Electrodes were attached to the
index and middle fingers of the left hand with medical tape, leaving the right hand free for
performing tasks. The average skin conductance response for each participant during the 5
minutes following the manipulation was normed by subtracting the average baseline skin
conductance for that participant during the 30 seconds prior to the moment the research assistant
began delivering the manipulation. This method was used instead of subtracting the average skin
conductance response from the resting measurement taken at the beginning of Part II of the study
because there was a large average difference in this resting measurement between rejection and
non-rejection (despite randomization of participants to condition). This pre-processing was
performed using the Python BioRead package (https://github.com/uwmadison-chm/bioread), and
the produced time series were compared to those created using the AcqKnowledge software to
ensure similarity. Mean centering was performed prior to testing the significance of interaction
effects (Iacobucci, Schneider, Popovich, & Bakamitsos, 2016). Statistical analysis was
performed using Python and R.
OTHER VARIABLE MEASURES
Relationships Questionnaire (RQ) – To assess self-described attachment style, participants
completed the RQ during part I of the study. The RQ consists of four descriptions of general
relationship styles (i.e., secure, preoccupied, dismissing, and fearful; e.g., “I am comfortable
without close emotional relationships” and “It is very important for me to feel independent and

18
self-sufficient, and I prefer not to depend on others or have others depend on me”). The RQ has
been validated with distinct profiles of interpersonal problems and has been shown to be reliably
correlated with family attachment ratings (Bartholomew & Horowitz, 1991). Participants first
placed a checkmark next to the relationship style that best describes them. They then rated the
degree to which each style described them on a 7-point scale (1 = Disagree strongly; 7 = Agree
strongly). The latter rating was not used for analyses in this study.
Trait Positive and Negative Affect Scale (T PANAS) – To assess trait affectivity, a 26-item T
PANAS was administered in part I of the study. The T PANAS asked participants to indicate to
what extent they had felt each emotion during the past few weeks. The standard 20-item PANAS
and six low arousal items (three positive: happy, satisfied, tranquil, and three negative:
frustrated, down, sad) were added to assess these additional dimensions of affect. Items were
rated on a 5-point scale (1 = Very slightly or not at all; 5 = Extremely). Previous research suggest
that the PANAS has good convergent and discriminant validity with other measures of
psychological functioning and distress (Watson et al., 1988). Although this instrument was
administered, the results were not used for subsequent analyses in this study.
Ten-Item Personality Interview (TIPI) – To assess personality style, the TIPI was administered
in part I of the study. The TIPI asked participants to rate the degree to which 10 short statements
(e.g., “Anxious, easily upset”) described them on a 7-point scale (1 = Disagree strongly; 7 =
Agree strongly). The measure has demonstrated convergent validity and shown a factor structure
consistent with the Five-Factor Model (FFM) of personality (Ehrhart et al., 2009). Although this
instrument was administered, the results were not used for subsequent analyses in this study.
Results
Manipulation Check
Based on the funnel debriefing, 12 participants stated that they did not believe the
manipulation and thus were eliminated from all subsequent analysis. Of the remaining 188
participants, 150 participants showed no signs of detecting the fact that the person they interacted
with was a confederate. 38 participants expressed uncertainty about whether deception was
involved, but were retained in all subsequent analyses since there was little evidence that their
observed reaction to the manipulation and responses to PANAS rejection items differed
substantially from the rest of the sample. In addition, during the funnel debriefing they did not

19
appear skeptical of the stated purpose of the study and generally indicated that they thought the
study was designed to measure what was described in the consent form.
Among the 188 participants whose data was used for the following analyses, ratings of
how well they thought the interaction had gone ranged from neutral to very well (0% very
poorly, 0% poorly, 2% neutral, 44% fairly well, and 54% very well). No participant indicated
that they were not looking forward to meeting the confederate again (73% indicated that they
were looking forward and 23% indicated that they were indifferent). The responses to these
items did not vary systematically as a function of the condition to which a participant was
randomly assigned, suggesting that an adequate expectation was created to meet again with the
confederate across both conditions (i.e., t(186) = 1.16, p = 0.25 and t(186) = -0.76, p = 0.45,
respectively). In order to verify that the statement by the RA in the experimental condition was
perceived as a rejection, we interspersed five rejection items in the state PANAS questionnaire
that was administered to participants before and after the manipulation. A two-way repeated
measures ANOVA indicated a significant Time x Condition interaction effect for change in self-
reported rejection, F(1,186) = 16.84, p = 0.00, partial eta2= 0.08. This led us to conclude that the
rejection manipulation was indeed successful and resulted in a medium effect.
Table 2 below presents descriptive statistics for participant characteristics across
conditions. The two groups did not differ significantly on any demographic or predictor variable.
Table 2. Participant Characteristics Across Conditions

Characteristic & Variable Reject Non-Reject χ2 t p

Gender 0.02 0.90
Female, freq (%) 68 (68.7) 61 (68.5)
Male, freq (%) 30 (30.3) 28 (31.5)
Ethnicity 19.37 0.06
White, freq (%) 58 (58.6) 44 (49.4)
Asian, freq (%) 23 (23.2) 29 (32.6)
Black or African American, freq (%) 8 (8.1) 5 (5.6)
Indian, freq (%) 6 (6.1) 5 (5.6)
Pacific Islander, freq (%) 3 (3.0) 2 (2.2)
Native American, freq (%) 1 (1.0) 0 (0.0)
More than one ethnicity, freq (%) 5 (5.1) 2 (2.2)
Other, freq (%) 8 (8.1) 2 (2.2)
Relationship status 4.02 0.57
Single, freq (%) 62 (62.6) 58 (65.2)
Casually dating, freq (%) 19 (19.2) 14 (15.7)
In a long-term relationship, freq (%) 17 (17.2) 19 (21.3)
Married, freq (%) 1 (1.0) 0 (0.0)
Relationship style 5.21 0.16

20
Table 2 (continued).

Secure, freq (%) 40 (40.4) 37 (41.6)

Fearful, freq (%)) 32 (32.3) 18 (20.2)
Preoccupied, freq (%) 16 (16.2) 16 (18.0)
Dismissing, freq (%) 11 (11.1) 18 (20.2)

Age, mean (SD) 19.2 (2.2) 19.0 (1.2) 0.44 0.66

ASA, mean (SD) 3.4 (0.5) 3.5 (0.5) -1.26 0.21
AB Schema, mean (SD) 2.5 (1.2) 2.3 (1.1) 0.94 0.35
MA Schema, mean (SD) 2.6 (1.2) 2.6 (1.1) 0.35 0.73
ED Schema, mean (SD) 1.6 (0.8) 1.5 (0.7) 0.97 0.33
SI Schema, mean (SD) 2.4 (1.1) 2.3 (1.1) 0.69 0.49
DF Schema, mean (SD) 1.8 (0.9) 1.6 (0.7) 1.49 0.14

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse
Schema, ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ
Defectiveness Schema. * p < .05.

Although ethnicity was near the nominal cutoff for significance, the difference appears due to the
small numbers of participants identifying ethnically as Black or African-American, Indian,
Pacific Islander, Native American, more than one ethnicity, or other.
Does secure base script predict the intensity of distress associated with social rejection better
than self-report schemas?
Based on the findings of Mclean, et al. (2014), we hypothesized that the ASA, Emotional
Deprivation Schema (ED Schema), Social Isolation Schema (SI Schema), and Defectiveness
Schema (DF Schema) would predict observed distress, but the Abandonment Schema (AB
Schema) and Mistrust/Abuse Schema (MA Schema) would not (Hypothesis 1). In addition, we
predicted that the ASA would be a better predictor of observed distress than any of the self-
report schemas (AB, MA, ED, SI, or DF Schemas) (Hypothesis 2). Finally, we predicted that the
ASA would be a better predictor of skin conductance response than any of the self-report
schemas (AB, MA, ED, SI, or DF Schemas) (Hypothesis 3).
Lead Researcher’s In-Person Coding of Observed Distress
A t-test of the lead researcher’s blind-to-condition, in-person coding of how upset
participants looked after the manipulation resulted in a statistically significant effect with
participants in the rejection condition appearing more upset that those in the non-rejection
condition, t(185) = 2.30, p = 0.02. It should be noted that for a few participants (approximately N
= 6) the blinding was compromised due to the need of the lead researcher to know the condition
that the participant was assigned to because of logistical challenges.

21
 Hierarchical regressions were conducted to assess whether any of the schema measures
accounted for the difference in how upset participants appeared across conditions. The condition
and schema variables were entered in step 1, and a Condition x Schema interaction was entered
in step 2. The statistics of interests (i.e., change in R2 values) and associated estimates for the
regressions are presented in Table 3.

Table 3. Brief Summary of Hierarchical Regression Analysis for Condition and Schema Variables
Predicting Blind, In-Person Coding of Observed Distress

Effect b SE b β step 1 R2 step 2 ∆R2

Condition x ASA -0.14 0.12 -0.11 0.03 0.01
Condition x AB Schema -0.01 0.12 -0.01 0.03 0.00
Condition x MA Schema 0.06 0.12 0.04 0.04 0.00
Condition x ED Schema 0.03 0.12 0.03 0.03 0.00
Condition x SI Schema 0.10 0.12 0.08 0.03 0.00
Condition x DF Schema -0.05 0.12 -0.04 0.04 0.00

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse Schema,
ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ Defectiveness
Schema. * p < .05.

As shown above, none of the schema measures significantly predicted the lead researcher’s in-
person rating of how upset participants looked after the manipulation. The full summary of these
regressions can be found in Appendix B.
Video Coding of Pre- and Post-Manipulation 1-Minute Clips of Observed Distress
Next, we examined the blind video coding of observed distress to test the hypotheses described
above. The rating of how upset participants appeared increased from pre- to post-manipulation in
both conditions as indicated by the means in Table 4 below.

Table 4. Full Summary of Repeated Measures Analysis of Variance for Condition Predicting
Change in Blind, Video Coding of Observed Distress

Effect Reject Non-Reject MS df F p ηp 2

Pre, mean (SD) 0.79 (0.53) 0.73 (0.52)

Post, mean (SD) 0.93 (0.54) 1.02 (0.49)

Time 4.43 1 18.32 0.00 0.10*

Time x Condition 0.01 1 0.02 0.88 0.00
Error 0.24 167

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse
Schema, ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ

22
Table 4 (continued).

Defectiveness Schema. * p < .05.

We conducted a repeated measures ANOVA to see whether the change in how upset participants
appeared differed by condition. In Table 4 above, the Time effect was significant (p = 0.00, ηp2 =
0.10*), but the Time x Condition interaction was not (p = 0.88, ηp2 = 0.00). This means that the
average increase in how upset participants appeared did not differ based on whether they
experienced the rejection or non-rejection manipulation. In fact, inspection of mean values in
Table 4 above indicates a surprising trend that the increase in how upset participants appeared
was greater in the non-rejection condition.
Repeated measures ANOVAs were conducted to further assess whether any of the
schema measures accounted for the change in how upset participants appeared across the two
conditions. These results are presented in Table 5 (full summary in Appendix B).

Table 5. Brief Summary of Repeated Measures Analysis of Variance for Schema Variables Predicting
Change in Blind, Video Coding of Observed Distress
2
Effect MS df F p Partial η
Time x Cond X ASA 0.27 1 1.11 0.29 0.01
Time x Cond X AB Schema 0.84 1 3.56 0.06 0.02
Time x Cond X MA Schema 2.00 1 8.87 0.00 0.05**
Time x Cond X ED Schema 0.09 1 0.37 0.54 0.00
Time x Cond X SI Schema 1.25 1 5.43 0.02 0.03*
Time x Cond X DF Schema 0.22 1 0.89 0.35 0.01

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse Schema, ED =
Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ Defectiveness Schema. * p < .05,
** p < .01.

Consistent with Hypothesis 1, the SI Schema (p = 0.02, ηp2 = 0.03*) significantly predicted
change in observed distress. Contrary to Hypothesis 1, the MA Schema (p = 0.00, ηp2 = 0.05**)
also significantly predicted change in observed distress and the AB Schema was close to the
nominal cutoff for statistically significance (p = 0.06, ηp2 = 0.02). Contrary to Hypothesis 2, the
ASA did not significantly predict change in observed distress by condition.
In order to interpret the nature of the AB, MA and SI Schema by Condition interactions,
we computed standardized expected values for the change in how upset participants appeared at
one standard deviation above and below the mean on each schema measure (as described by
Cohen & Cohen, 1983). The results are presented in Figure 2 below.

23
 Blind Video Coding of How Upset Blind Video Coding of How Upset
Participant Appeared Participant Appeared
Increase in How Upset

Increase in How Upset

0.4 1

0.2 0.5
0
0
Reject Non-Reject
Reject Non-Reject -0.5
Condition Condition

Low Abandonment High Abandonment Low Mistrust/Abuse High Mistrust/Abuse

Blind Video Coding of How Upset

Participant Appeared
Increase in How Upset

0.5

0
Reject Non-Reject
-0.5
Condition

Low Social Isolation High Social Isolation

Figure 2. Change in how upset participants appeared as predicted by Condition and AB Schema (top left) / MA
Schema (top right) / SI Schema (bottom). Individuals low on these schemas are in blue and those high on these
schemas are in red.

The graphs in Figure 2 indicate a similar pattern for all three schemas. In particular, participants
scoring high on the AB, MA, and SI Schemas demonstrated an increase in how upset they
appeared in the rejection condition compared to the non-rejection condition. Conversely,
participants scoring low on these schemas demonstrated a decrease in how upset they appeared
in the rejection condition compared to the non-rejection condition.
There are two main conclusions to draw from these results: (a) explicit self-report
schemas (AB, MA, and SI) exhibited better predictive validity for observed distress than an
implicit measure (the ASA); and (b) these explicit schemas predicted large differences in
observed distress in response to the non-rejection condition but not in response to rejection.
Skin Conductance Response
We predicted that the ASA would be a better predictor of skin conductance response than
any of the self-report schemas (AB, MA, ED, SI, or DF Schemas) (Hypothesis 3). In order to test

24
this hypothesis, we recorded skin conductance response for 2 minutes prior to the manipulation
and 5 minutes after the manipulation. In the following analyses, we excluded 27 recordings that
had short or very low voltage recordings, 26 in which there was an error in marking when the
manipulation was delivered, and 5 for which we did not have corresponding ASA or YSQ data.
The average skin conductance response for each participant during the 5 minutes following the
manipulation was normed by subtracting the average baseline skin conductance for that
participant during the 30 seconds prior to the moment the research assistant began delivering the
manipulation. This method was used instead of subtracting the average skin conductance
response from the resting measurement taken at the beginning of Part II of the study because
there was a large average difference in this resting measurement between rejection and non-
rejection (despite randomization of participants to condition). The graph in Figure 3 below
depicts the average participant-normed skin conductance response by condition. Since the
manipulation took approximately 20 seconds to verbally state, the x-axis begins at -20 seconds
and the delivery of the manipulation is complete at t = 0 seconds.

Figure 3. Change in average skin conductance for the five minutes following the manipulation, graphed by
condition (reject in blue and non-reject in orange). The shaded regions represent 95% confidence intervals.

As can be seen above, both conditions witnessed a peak in skin conductance response at t = 0
seconds. Based on inspection of the graph, we determined that the peak response in the rejection
condition (blue) was more than double the peak in the non-rejection condition (orange). The
effect appeared substantial throughout the first 30 seconds following the manipulation,
suggesting that the manipulation was indeed effective.

25
 Next, we conducted a hierarchical regression using the first 30 seconds of skin
conductance response to see whether the difference between conditions was further predicted by
the ASA. The condition and ASA variables were entered in step 1, and a Condition x ASA
interaction was entered in step 2. The estimate of interest (i.e., change in R2) and associated
statistics for the regression is presented in Table 6.
Table 6. Full Summary of Hierarchical Regression Analysis for Condition and ASA Predicting Skin
Conductance Response

Variable b b 95% CI R2 ∆R2

[UL, LL]
Step 1
Constant 0.08 [-0.39, 0.54]
Condition 1.15** [0.51, 1.80]
ASA 0.06 [-0.59, 0.72] 0.09**
Step 2
Constant 0.09 [-0.37, 0.55]
Condition 1.15** [0.52, 1.79]
ASA -0.74 [-1.71, 0.23]
Condition x ASA 1.46* [0.16, 2.76] 0.03*

Note. ASA = Adult Script Assessment. LL and UL indicate the lower and upper limits of a confidence interval,
respectively. * p < .05. ** p < .01.

Consistent with Hypothesis 3, there was a significant Condition x ASA interaction (b = 0.72, p <
.05, ∆R2 = 0.03*), and none of self-report schemas significantly predicted the difference in skin
conductance response across conditions (see Table 14 in Appendix B for full summary). The
nature of the Condition x ASA interaction is depicted in Figure 4 below.

SCR by Condition and ASA

2.5
Mean SCR 0 to 30 seconds
following manipulation

2
1.5
1
0.5
0
Reject Non-Reject
-0.5
Condition
-1
Low Security ASA High Security ASA

Figure 4. The Condition by Adult Script Assessment (ASA) interaction. The graph on the left depicts t-values
associated with the difference between the two conditions for the first 30 seconds following the manipulation (the
dotted line represents the cutoff for statistical significance at p = 0.05). The graph on the right depicts the average
skin conductance response over this time period by condition and high/low security ASA.

26
The dotted line in the graph on the left in Figure 4 represents the cutoff for statistical
significance. As demonstrated by the graph on the right in Figure 4, the nature of the interaction
was opposite to what we predicted. That is, individuals with high levels of secure base script
knowledge (blue) showed a large increase in skin conductance response in the rejection
condition compared to non-rejection, whereas those with low levels of secure base script
knowledge (orange) showed less response to the rejection condition and more of a response to
the non-rejection (p = 0.028, N = 130).
Overall, this means that Hypothesis 3 was supported and the implicit ASA measure
emerged a better predictor of immediate sympathetic response than did any explicit self-report
schema. The nature of the interaction was such that those low in secure base script appeared
more reactive to the non-rejection condition than those high in secure base script. Conversely,
those low in secure base script appeared less reactive to the rejection condition than those high in
secure base script.
Does secure base script predict shifts in phenomenological perspective associated with social
rejection based on relationship style?
Regardless of relationship style, we hypothesized that individuals low on secure base
script knowledge would be more likely to express negative affective expectations of others
following the manipulation in the rejection condition compared to the non-rejection (Hypothesis
4). By contrast, we predicted that only those who endorsed an anxious-avoidant (i.e., dismissing)
relationship style would display anger (Hypothesis 5), and only those who endorsed an anxious-
resistant (i.e., preoccupied) relationship style would show a decrease in self-esteem following
rejection (Hypothesis 6). We made no predictions for individuals high on both avoidance and
resistance (i.e., a fearful relationship style).
The change in SCORS variables across time and by condition is illustrated in the mean
ratings for these variables presented in Table 7. Somewhat counter-intuitively, the Affective
Quality of Representations and Self-Esteem variables increased from pre- to post-manipulation
in the rejection condition. This means that participants exhibited more positive expectations for
relationships and viewed themselves more positively after being rejected. By contrast, the
Experience and Management of Aggressive Impulses variable decreased from pre- to post-
manipulation in both the rejection and non-rejection conditions. This means that participants
exhibited less effective control over their anger regardless of the condition. We conducted

27
repeated measures ANOVAs to determine if the Time or Time x Condition effects were
significant.

Table 7. Full Summary of Repeated Measures Analysis of Variance for Condition Predicting
Change in Social Cognition and Object Relations Scales (SCORS)

Affective Quality of Representations

Effect Reject Non-Reject MS df F p ηp 2

Pre, mean (SD) 3.5 (0.36) 3.5 (0.28)

Post, mean (SD) 3.6 (0.44) 3.5 (0.48)

Time 0.43 1 4.55 0.03 0.03*

Time x Condition 0.19 1 0.19 0.16 0.01
Error 0.09 180
Experience and Management of Aggressive Impulses
Effect Reject Non-Reject MS df F p ηp 2

Pre, mean (SD) 3.8 (0.21) 3.8 (0.20)

Post, mean (SD) 3.7 (0.35) 3.6 (0.34)

Time 0.71 1 13.03 0.00 0.07*

Time x Condition 0.08 1 1.54 0.22 0.01
Error 0.05 180
Self-Esteem
Effect Reject Non-Reject MS df F p ηp 2

Pre, mean (SD) 3.9 (0.11) 4.0 (0.11)

Post, mean (SD) 4.0 (0.15) 3.9 (0.16)

Time 0.01 1 0.76 0.38 0.00

Time x Condition 0.05 1 3.45 0.07 0.02
Error 0.02 180

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse
Schema, ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ
Defectiveness Schema. * p < .05, ** p < .01.

As can be seen in Table 7 above, although the Time effect was significant for Affective Quality
of Representations and Experience and Management of Aggressive Impulses, none of the Time x
Condition effects were statistically significant. This means that these two SCORS variables
changed significantly from pre- to post-manipulation, but the change in these variables was not
moderated by the condition to which participants were assigned.
Since our hypotheses were specifically about the interaction between relationship style
and ASA on the SCORS outcome variables, we conducted repeated measure ANOVAs with

28
separate contrasts for each of the four relationship styles (i.e., secure, dismissing, preoccupied, or
fearful) endorsed by participants on the Relationship Questionnaire (RQ). For instance, for the
secure contrast, participants endorsing a secure relationship style were dummy coded 1 and
participants endorsing any other relationship style (i.e., dismissing, preoccupied, or fearful) were
coded 0. In Table 8 below, the results are presented for the four relationship style contrasts with
change in Affective Quality of Representations as the outcome variable.
We hypothesized that the ASA would predict change in Affective Quality of
Representations for all four of the relationship styles (Hypothesis 4), however, the results in
Table 8 indicate that only the secure relationship style had a statistically significant Time x
Condition x ASA association.

Table 8. Brief Summary of Repeated Measures Analysis of Variance for Adult Script Assessment (ASA)
Predicting Change in Affective Quality of Representations by Relationship Style
2
Effect MS df F p Partial η
Time x Cond x Secure x ASA 0.73 1 8.12 0.01 0.05*
Time x Cond x Dismissing x ASA 0.15 1 1.60 0.21 0.01
Time x Cond x Preoccupied x ASA 0.13 1 1.35 0.25 0.01
Time x Cond x Fearful x ASA 0.04 1 0.37 0.54 0.00

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

This interaction was in the medium effect range. This means that the ASA significantly predicted
the pre/post change in positive expectations of others as measured by the SCORS, but only for
individuals who self-reported as having a secure relationship style.
In order to further explore the nature of the significant interaction for the secure contrast,
we graphed standardized expected values for secure versus non-secure individuals by high/low
security ASA and condition.

29
 Affective Quality of Representations

Participants Self-Reporting a Participants Self-Reporting a

Secure Relationship Style Non-Secure Relationship Style

Increase in Affective Quality of

Increase in Affective Quality of

0.3 0.2

Representations
Representations

0.15
0.2
0.1
0.1
0.05
0 0
Reject Non-Reject Reject Non-Reject
Condition Condition

High Security ASA Low Security ASA High Security ASA Low Security ASA

Figure 5. Increase in Affective Quality of Representations as predicted by Condition and Adult Script Assessment
(ASA). Participants who self-reported a secure relationship style on the Relationships Questionnaire are depicted in
the graph on the right, and those self-reporting a non-secure relationship style (i.e., preoccupied, dismissing, or
fearful) are depicted in the graph on the left. Individuals low on secure base script knowledge are depicted in blue
and those high on secure base script knowledge are in red.

The increase in Affective Quality of Representations is depicted in Figure 5 above. The general
pattern seen is that the increase in positive expectations of others from pre- to post-manipulation
was greater in the rejection condition compared to non-rejection. The only instance in which the
pattern was reversed was when individuals’ explicit self-report of having a secure relationship
style contradicted the fact that they exhibited low levels of implicit secure base script knowledge
(i.e., the red line in the graph on the left). This means that when a participant exhibited this kind
of explicit-implicit contradiction, they showed much less of an increase in positive expectations
of others following rejection.
In Table 9 below, the results are presented for the four relationship style contrasts with
the change in Experience and Management of Aggressive Impulses as the outcome variable. We
hypothesized that the ASA would predict change in Experience and Management of Aggressive
Impulses for the dismissing relationship style (Hypothesis 5), however, again only the secure
relationship style had a statistically significant Time x Condition x ASA association.
Table 9. Brief Summary of Repeated Measures Analysis of Variance for Adult Script Assessment (ASA)
Predicting Change in Experience and Management of Aggressive Impulses
2
Effect MS df F p Partial η
Time x Cond x Secure x ASA 0.30 1 5.95 0.02 0.03*

30
Table 9 (continued).

Time x Cond x Dismissing x ASA 0.01 1 0.15 0.70 0.00

Time x Cond x Preoccupied x ASA 0.11 1 2.03 0.16 0.01
Time x Cond x Fearful x ASA 0.05 1 0.87 0.35 0.01

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

This interaction was in between a small and medium effect. This means that the ASA
significantly predicted the pre/post change in the ability of participants to manage their anger as
measured by the SCORS, but only for individuals who self-reported as having a secure
relationship style.
In order to further explore the nature of the significant interaction, we graphed
standardized expected values for secure versus non-secure individuals by high/low security ASA
and condition. The Experience and Management of Aggressive Impulses variable decreased
from pre- to post-manipulation in both conditions, indicating that participants were less effective
at managing their anger after experiencing either manipulation. Figure 6 below depicts this
decrease in Experience and Management of Aggressive Impulses.

Experience and Management of Aggressive Impulses

Participants Self-Reporting a
Management of Agressive Impulses

Participants Self-Reporting a
Management of Agressive Impulses

Secure Relationship Style Non-Secure Relationship Style

Decrease in Experience and
Decrease in Experience and

0.05 0.05
0 0
Reject Non-Reject Reject Non-Reject
-0.05
-0.05
-0.1
-0.15 -0.1

-0.2 -0.15
Condition Condition

High Security ASA Low Security ASA High Security ASA Low Security ASA

Figure 6. Decrease in and Management of Aggressive Impulses as predicted by Condition and Adult Script
Assessment (ASA). Participants who self-reported a secure relationship style on the Relationships Questionnaire are
depicted in the graph on the right, and those self-reporting a non-secure relationship style (i.e., preoccupied,
dismissing, or fearful) are depicted in the graph on the left. Individuals low on secure base script knowledge are
depicted in blue and those high on secure base script knowledge are in red.

31
The decrease in Experience and Management of Aggressive Impulses is depicted in Figure 6
above. The general pattern seen in the graphs above is that the decrease in the ability to manage
anger from pre- to post-manipulation was greater in the non-rejection condition compared to
rejection. The only instance in which the general pattern was reversed was again when
individuals’ explicit self-report of having a secure relationship style contradicted the fact that
they exhibited low levels of implicit secure base script knowledge (i.e., the red line in the graph
on the left). This means that when a participant exhibited this kind of explicit-implicit
contradiction, they showed much more of a decrease in ability to effectively manage anger
following rejection.
In Table 10 below, the results are presented for the four relationship style contrasts with
the change in Self-Esteem as the outcome variable. We hypothesized that the ASA would predict
change in Self-Esteem only for the preoccupied relationship style (Hypothesis 6). Contrary to our
prediction, none of the four relationship styles had a statistically significant Time x Condition x
ASA association.

Table 10. Brief Summary of Repeated Measures Analysis of Variance for Adult Script Assessment (ASA)
Predicting Change in Self-Esteem

2
Effect MS df F p Partial η
Time x Cond x Secure x ASA 0.02 1 1.17 0.28 0.01
Time x Cond x Dismissing x ASA 0.00 1 0.02 0.89 0.00
Time x Cond x Preoccupied x ASA 0.00 1 0.21 0.65 0.00
Time x Cond x Fearful x ASA 0.02 1 1.19 0.28 0.01

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

This means that the ASA did not significantly predicted the pre/post change in self-esteem as
measured by the SCORS for any of the four relationship styles. The full summary statistics of
these repeated measures ANOVAs can be found in Appendix B.
Taken together, our results contradicted Hypotheses 4, 5, and 6. Although low secure
base script did predict shift in phenomenological perspective, it did so for participants self-
reporting a secure relationship style. These participants were unique, exhibiting less of an
increase in positive expectations of others and more loss of control over anger following the
rejection manipulation. All other participants had this reaction to the non-rejection manipulation.

32
 Discussion
Although cross-sectional studies have shown that both implicit and explicit measures of
schemas are related to signs and symptoms of psychopathology, the limitation of these designs is
that they tell us nothing about the dynamic role of memory structures in the genesis of distress. In
order to understand how such schemas function in directing mood and behavior, the present
study utilized a social rejection paradigm to investigate the causal relationship between
abandonment schema activation and moment-to-moment expression of distress.
The first contribution of this study was to shed light on the differential predictive utility
of implicit versus explicit schema measures in explaining distress. Explicit self-report schemas
were better predictors of change in observed distress than was implicit secure base script
knowledge. Based on previous correlations found between the ASA and various YSQ subscales
(McLean et al., 2014), we hypothesized that the emotional deprivation (ED Schema), social
isolation (SI Schema), and defectiveness (DF Schema) subscales of YSQ would be associated
with observed distress, but the abandonment (AB Schema) and mistrust/abuse (MA Schema)
subscales would not (Hypothesis 1). Curiously, the ratings of how upset participants appeared
increased significantly from pre- to post-manipulation in both conditions. In fact, the increase in
how upset participants appeared was slightly higher in the non-rejection compared to the
rejection condition (see Table 4). Since participants self-reported feeling more rejected in the
rejection condition, it does not appear that the change in observed distress tracked with
participant self-report.
Findings for Hypothesis 1 were mixed. As predicted, the SI Schema was significantly
associated with the change in how upset participants appeared across conditions. However
contrary to the correlations found in McLean et al. 2014, participants high in the MA and AB
Schemas also showed an increase in how upset they appeared in the rejection condition
compared to non-rejection. Furthermore, these three explicit schemas seemed to capture large
differences in observed distress in response to non-rejection rather rejection. Taken together,
these results suggest that the SI, MA, and AB schemas are potentially useful explicit measures
that predict differences in observed change in facial/non-verbal behavior when individuals
experience a non-rejection stimulus. This finding is consistent with the idea that the YSQ serves
as a useful measure of behavioral responses that are observed by the individual or others, and
therefore captures behavioral tendencies that are more likely to be available for self-report. As

33
illustrated in the increased predictive utility of theses schemas in the non-rejection condition, the
association between schema content and the determinants of behavior in actual meaningful
environments are complex.
By contrast, the ED and DF schema subscales and the ASA (Hypothesis 2) did not
demonstrate predictive utility for change in observed distress across conditions. Our hypotheses
for this outcome measure were based on correlations between the ASA and the YSQ schema
sub-scales found in a cross-sectional study of 146 undergraduates conducted by McLean et al.
(2014). Of note, the ASA measure in McLean et al.’s study assessed secure base script
knowledge using undergraduates’ descriptions of three of their current relationships rather than
the standard word prompt method used in this study. It is possible that these different
methodologies contributed to the different pattern of results. Alternatively, the difference in our
findings may suggest caution in assuming that correlations between explicit and implicit
measures necessarily imply that they will similarly predict behavioral outcomes in meaningful
situational contexts.
Whereas explicit self-report schemas were better predictors of observed distress, the
opposite was true for implicit distress (Hypothesis 3). That is, an implicit measure (i.e., secure
base script knowledge) was a better predictor of participants’ implicit response (i.e., autonomic
sympathetic reactivity) than any explicit self-report schema. Although Hypothesis 3 was
confirmed, surprisingly the nature of the interaction was such that participants who scored high
on secure base script knowledge displayed a greater change in sympathetic response in the
rejection condition compared to non-rejection. Participants scoring low on secure base script
appeared to have a “blunted” physiological response in the rejection condition and displayed a
greater change in sympathetic response to the non-rejection stimulus.
How can we understand this difference in physiological reactivity? Since skin
conductance response is closely linked to autonomic emotional and cognitive processing, it can
capture implicit emotional responses that occur without conscious awareness (Braithwaite,
Watson, Jones, & Rowe, 2015). Although all participants (regardless of ASA score) self-reported
feeling more rejected in the rejection condition,2 it was only those with a high degree of secure
base script knowledge that had a congruent autonomic-affective response. An intriguing

2
A repeated measures ANOVA with ASA predicting change in self-report rejection items from pre- to post-
manipulation was non-significant, F(183) = 0.02, p = 0.89.

34
possibility is that Low Security ASA participants may have been more likely to expect rejection
based on past experience and may have thus been less reactive to rejection in the moment. This
might also explain why they were more reactive to the non-rejection manipulation. In this
condition, the meaning of the statement “there is not enough time for the second interaction”
could result in a wide range of possibilities for interpretation (including that the statement was
meant to conceal that the participant was rejected by the confederate). In this sense, the stimulus
in the non-rejection condition could more accurately be characterized as ambiguous as opposed
to non-rejecting.
Understood within this context, low secure base script individuals may be poised for the
possibility of rejection at any moment. For instance, take the case of an individual who displays
borderline features including “frantic attempts to avoid real or imagined abandonment.” A
clinical scenario illustrating this feature might include a situation in which a therapist states that
there are five minutes left to the session, and the client immediately brings up suicidal ideation
from the week (presumably as an attempt to secure the care and attention of therapist). It is the
ambiguous nature of the situation that leads the individual to impose an interpretation that the
therapist is abandoning them. This clinical scenario is consistent with the pattern of skin
conductance response of the low ASA participants (orange line) in figure 1. These individuals
show an increase in emotional arousal in response to an ambiguous stimulus (i.e., the non-
rejection condition) as opposed to an instance of direct threat (i.e., the rejection condition).
Alternatively, perhaps those low in secure base script had some kind of implicit
mechanism for reducing negative affective experience. For instance, Low Security ASA
participants may have approached the confederate with less emotional investment in order to
avoid the possibility of future distress associated with abandonment or rejection. Such a
protective strategy could be understood as an adaptation, based on prior learning history, to
minimize the impact of threats to self-stability. Although this strategy may appear beneficial in
the context of the experiment, it could prove to be a counterproductive barrier when attempting
to form close relationships in daily life. Another possibility is that the blunted autonomic
response observed meant that these individuals could not experience an emotional reaction, but
still felt the impact of the manipulation—after all, they did report feeling rejected. If distress
associated with rejection did still affect these individuals in some way, perhaps it was expressed
at a later time or in a different context (e.g., exploding at a romantic partner after the study).

35
 The second contribution of this study was to examine the relationship between secure
base script knowledge and shifts in phenomenological perspective following rejection. More
specifically, we asked, is the ASA a significant predictor of shift in one’s interpretive and
affective world following rejection? Such a finding could clarify whether schema activation
momentarily “colors” an individual’s perception of reality, leading individuals to behave
differently in time based on specific situational events. In order to assess this kind of shift in
phenomenological perspective, we measured changes in participants’ responses to TAT cards
using the affective quality of representations (AFF), experience and management of aggressive
impulses (AGG), and self-esteem (SE) subscales of the Social Cognition and Object Relations
Scale (SCORS). We hypothesized that individuals low on secure base script knowledge would be
more likely to express negative affective expectations of others following the manipulation in the
rejection condition compared to the non-rejection (Hypothesis 4). By contrast, we predicted that
only those who endorsed an anxious-avoidant (i.e., dismissing) relationship style would display
anger (Hypothesis 5), and only those who endorsed an anxious-resistant (i.e., preoccupied)
relationship style would show a decrease in self-esteem following rejection (Hypothesis 6).
Somewhat counter-intuitively, participants’ positive expectations of others (AFF) and
self-esteem (SE) as measured by the SCORS increased from pre- to post-manipulation in the
rejection condition (see means in Table 7). Given that participants self-reported as feeling more
rejected in this condition, one possibility is that this shift in phenomenological perspective
represents a kind of affective-buffering following a distressing event. Nonetheless, participants’
ability to manage aggressive impulses (AGG) decreased from pre- to post-manipulation in both
conditions. This suggests that despite potential attempts at buffering their experience, the
manipulation did result in some loss of control of anger in both conditions.
Contrary to our predictions, the ASA was significantly associated with shifts in
phenomenological perspective only for participants who self-reported as having a secure
relationship style on the RQ. In particular, those with low secure base script knowledge who
thought they had a secure relationship style exhibited less of an increase in positive expectations
of others and more loss of control over anger following the rejection manipulation. It appears
that participants with this kind of explicit-implicit contradiction (i.e., who lacked insight into
their reactivity) were more vulnerable to the rejection manipulation. In this sense, the ASA
seems to provide a form of predictive utility beyond categorical attachment designations. It

36
distinguishes individuals who self-report as secure, but nonetheless appear to have a latent
vulnerability to rejection. This finding would be consistent with the idea that without a
developed secure base script, individuals may lack the internal resources to buffer distress and
soothe themselves in the face of rejection and/or abandonment. When their phenomenological
world shifts, they may have less capacity to observe their reactions and self-regulate.
As a whole, the present study aimed to investigate the dynamic experiential and
behavioral changes associated with abandonment schema activation. The strengths of the study
include the use of both implicit and explicit schema measures and multiple kinds of outcome
measures (i.e., self-report instruments, video capture of observed distress, psychophysiological
recordings, and projective measures of momentary shifts in phenomenological perspective).
Furthermore, since participants were randomized to rejection and non-rejection conditions,
causal inferences could be drawn regarding the nature of the responses to each manipulation.
Limitations include the fact that although we reached the goal for overall sample size based on
an a priori power analysis, the number of participants with usable skin conductance data was
smaller than expected. In addition, the inter-rater reliability for ratings of observed distress and
the self-esteem SCORS variable were lower than anticipated. Finally, although use of non-
clinical undergraduate sample was a useful starting point for investigating abandonment schema
activation, generalization to populations with significant psychopathology will require future
studies. Overall, the study sheds light on how implicit representational processes have causal
significance for the moment-to-moment experience and behavior of individuals in meaningful
situation contexts that involve the possibility of feeling rejected.

37
 CHAPTER 3: IDENTIFYING INFORMATION PROCESSING ELEMENTS OF AN
ABANDONMENT SCHEMA
Introduction
Study 2 further assessed the information processing (self-other) meaning elements that
may be associated with the implicit memory of abandonment distress depicted in Figure 1
(Chapter 1). More specifically, it tested the conjecture that individuals who have had repeated
experiences of their primary caretaker as accessible and attentive form representations of
themselves as loved and others as responsive, whereas those that do not have repeated
experiences of their primary caretaker as accessible and attentive form representations of
themselves as unworthy of care and others as rejecting and unresponsive. The proposed
organization of meaning elements was investigated through a modified Rejection Stroop task
(Baldwin & Sinclair, 1996) in which participants were asked to keep a prime phrase in mind and
repeated it following the color naming of a target word (Segal, Gemar, Truchon, Guirguis, &
Horowitz, 1995). The relatedness of the prime and target was the variable of interest. Slower
color-latencies for matching prime-target pairs versus non-matching pairs supports the idea that
when both elements belong to the same cognitive representation, exposure to the prime
automatically increases accessibility of the target word’s meaning, which interferes with naming
the color in which it is presented.
We designed the Abandonment Stroop to examine whether access to abandoned self-
states can be enhanced by prior presentation of personalized prime slides that depicted phrases
and images related to each participant’s past experiences of abandonment. If color naming of
abandonment target words was slower when primed by abandonment primes than by neutral
primes, it would suggest that the specific interference observed in individuals with a lack of
consolidated secure base script is mediated by a representation in which this information is
organized in this manner. Information that is not organized in this manner (e.g., abandonment
information for those with high secure base script knowledge) should not lead to increased
interference. We hypothesized that individuals with low secure base script knowledge would
show the longest color-naming latencies when an abandoning self-relevant prime slide was
paired with an abandonment target word.
To summarize, the goal of Study 2 was to test the hypothesis that individuals who had
less consolidated secure-base scripts were likely to process information relevant to past

38
abandonment experiences differently than individuals who had more consolidated secure base
scripts.
Method
Design
There was one between-subjects variable (score on the ASA secure base script measure)
and two within-subjects variables: prime slides (2 types: abandonment vs. neutral) and target
word slides (2 types: abandonment vs. neutral).
Each trial consisted of presentation of an initial prime slide (image with phrase
superimposed in one of four corners), followed by a target word slide (with word presented in
red, yellow, blue, or green; see Figure 7 below). The participant was asked to keep the phrase
from the prime slide “in mind,” and was told to state the color of the target word followed by the
prime phrase in order to demonstrate they had “kept the phrase in mind.”
We manipulated the type of prime slide (abandonment vs. neutral) and target word slide
(abandonment vs. neutral) to give four types of trials: (a) AA – abandonment prime and
abandonment target, (b) NA – neutral prime and abandonment target, (c) AN – abandonment
prime and neutral target, and (d) NN – neutral prime and neutral target.

39
Figure 7. All trials began with a fixation cross followed by a blank screen, then participants viewed an abandonment
prime slide with a personalized phrase (depicted) or a neutral prime slide with a neutral phrase (Note: the phrase
appeared in one of four corners of the image with a delay of 0.2s). The prime slide was followed by a blank screen
and then a target word slide that was either an abandonment word (depicted) or a neutral word. The participant said
the color of the word (“yellow”) followed by the phrase (”Mom does not respond”). Each trial ended with a blank
screen.

Participants
A power analysis was conducted to determine the necessary sample size to detect a
medium effect (Cohen’s f = 0.25, α = 0.05, power = 0.80; total sample size = 128). Based on this
analysis and feasibility, 209 participants were eventually recruited for the study. In order to
increase the likelihood that participants would display a sufficient range of scores on our
measure of secure base script knowledge (i.e., the ASA), we planned to recruit 80% of the
sample as new participants from the UIUC course credit subject pool and aimed to recruit the
other 20% of the sample from Study 1 by emailing participants who scored above 4 and below 3
on the ASA (these thresholds were based on inspection of a histogram of ASA scores for
participants from Study 1 which indicated that about 80% of sample was between these values).

40
These participants were paid $15 in lieu of subject pool credit. Unfortunately, only 5 participants
from Study 1 responded and completed the present study. Inclusion criteria for the study
included agreement through written informed consent to participate in a one-hour study and
documentation that the participant was over the age of 18. There were no explicit exclusion
criteria for the study.
The undergraduate participants ranged from age 18-26 (one person was age 39) and 79%
identified as female (20% male). Participant’s year in college ranged from freshman to seniors,
41% freshmen, 32% sophomore, 17% juniors, and 9% seniors. In terms of ethnic background,
18% identified as Hispanic or Latinx and 82% identified as Not Hispanic or Latinx. In terms of
racial background, 47% identified as Caucasian, 24% Asian or Asian American, 11% as Black or
African American, 7% Mixed Race, 6% Indian Subcontinent, 1% Native American, less than 1%
Pacific Islander, and 5% Other. 85% were native English speakers, and 15% were competent in
English.
Stimulus Material Generation & Construction for the Experimental Trials
Image selection. The two neutral images used on the prime slides were the two neutral pictures
from the Adult Attachment Picture (AAP) system (see first two slides in AAP images in
Appendix A). The two abandonment images were chosen by each participant in Part 1 of the
study from the eight abandonment pictures in the AAP system (see remaining eight slides in
Appendix A). In part 1, each participant was told, “Everyone has past experiences in which they
felt rejected or abandoned. On the next screen you will see eight pictures that depict distressing
situations. Look through the pictures and choose two pictures that you associate most with
your own personal experiences of rejection/abandonment. After you have chosen two
pictures, write a short paragraph (3-5 sentences) describing the memories, bodily sensations,
thoughts, and feelings that you remember during these event(s). PLEASE DO NOT DESCRIBE
THE PICTURE. ONLY USE THE PICTURE AS A PROMPT/CUE TO WRITE ABOUT
YOUR OWN PERSONAL MEMORIES.” This was done to elicit an emotional story that was
more likely to become associated with each image. All images used on the prime slides were line
drawings from the AAP that are similar in terms of size and thematic content.
Phrase selection & prime slide construction. In order to create neutral prime phrases to
superimpose on each prime slide image, we conducted a pilot study in which 20 undergraduates
rated 16 neutral phrases designed to describe the two neutral images in the AAP. To minimize

41
variability in length and content, all phrases were 3-5 words that describe non-relational aspects
of the pictures. In order to validate the neutrality of the phrases, each phrase was rated in terms
of the degree to which it conveyed a sense of abandonment (not at all, slightly, moderately,
extremely) (e.g., “the ball is bouncing,” “the hat is on backward,” “one child is sitting,” “the bed
is new,” and “the boy is young,”). Out of the phrases that were validated as “not at all
abandoning,” three neutral phrases were chosen for the first image and three for the second. The
six unique neutral prime slides were constructed by superimposing each of three prime phrases
individually on the two neutral images.
In order to create abandonment prime phrases to superimpose on each prime slide image
chosen by the participant, we added to the pilot study described above 16 abandonment phrases
designed to describe the eight abandonment images in the AAP. The phrases had blanks for the
insertion of proper nouns by participants that allowed the phrases to be personalized for their
own experiences of abandonment. To minimize variability in length and content, all phrases
were 3-5 words and each described relational (two-person) aspects of the pictures that were
synonyms for abandonment constructed from the abandonment items of the Young Schema
Questionnaire (i.e., they: “_____ rejected me,” “_____ hurt me,” “_____ does not respond,”
“_____ are leaving me,” “_____ does not care,” and “_____ dislikes me”). In order to validate
the phrases, each was rated in terms of the degree to which the phrase conveyed a sense of
abandonment (not at all, slightly, moderately, extremely). The 10 phrases that were validated as
most abandoning were presented to participants in part I of the study. Participants chose six to
personalize and matched three to their first abandonment image and three to their second
abandonment image. The six unique abandonment prime slides were constructed by
superimposing each of three prime phrases individually on the two abandonment images.
Target word selection. The abandonment target word slides consisted of 12 abandonment words
from a previous rejection Stroop task (Dandeneau & Baldwin, 2004). The abandonment words
from this study were: “unwanted,” “ignored,” “rejected,” “disliked,” “shunned,” “rebuffed,”
“neglected,” “excluded,” “avoided,” “isolated,” “condemned,” and “disapproved.” The neutral
target word slides consisted of 12 neutral words taken from the pilot study in which 20
undergraduates rated 16 neutral words that are non-relational in content. To minimize variability
in length and usage, all words were similar in length to the abandonment words and described
objects (e.g., “furniture,” “building” “elevator,” “stoplight,” “vehicle,” “apartment,” “potato,”

42
“restaurant,” “elephant,” “cucumber,” “pencil,” “button,” “umbrella,” “television,”
“headphones,” “radio”). In order to validate the neutrality of the words, each was rated in terms
of the degree to which it conveyed a sense of abandonment (not at all, slightly, moderately,
extremely). Out of the words that were validated as “not at all abandoning,” twelve were chosen.
Prime slide—Target word trial construction. Each of the 6 abandonment prime slides were
randomly paired with 4 of the 12 abandonment target word slides for a total of 24 trials of the
first trial type (AA). Each of the 6 neutral prime slides were randomly paired with 4 of the 12
abandonment target word slides for a total of 24 trials of the second trial type (NA). Each of the
6 abandonment prime slides were randomly paired with 4 of the 12 neutral target word slides for
a total of 24 trials of the third trial type (AN). Each of the 6 neutral prime slides were randomly
paired with 4 of the 12 neutral target word slides for a total of 24 trials of the fourth trial type
(NN).
As depicted in Figure 1, for each experimental trial the prime phrase randomly appeared
superimposed on one of the four corners of the picture with a 0.2s delay so that participants
needed to pay attention to both image and phrase on every trial. The colors of target words (red,
yellow, blue, or green) were randomized with no two consecutive trials consisting of the same
color. Based on the specifications above, there were a total of 96 trials (24 trials per 4 trial types:
AA, NA, AN, NN). Each of six abandonment prime slides were presented 4 times in the AA
trials and 4 times in the AN trials for a total of 8 times over the course of the task. Each of six
neutral prime slides were presented 4 times in the NA trials and 4 times in the NN trials for a
total of 8 times over the course of the task. Each of twelve abandonment target words were
presented 2 times in the AA trials and 2 times in the NA trials for a total of 4 times over the
course of the task. Each of twelve neutral target words were presented 2 times in the AN trials
and 2 times in the NN trials for a total of 4 times over the course of the task.
Since each of the 24 total prime slides was always paired with 4 different target word
slides over the course of the task, the total number of trials was 96 (24 * 4). These 96 trials were
randomly ordered into 12 blocks of 8 trials each such that there was two of each trial type (AA,
NA, AN, NN) per block with no two consecutive trials of the same type (e.g., AA).
Experimental Procedure
Participants completed Part I of the study online. Part I took 25 minutes and began with
an informed consent form describing the entire study. If the participant chose to participate, they

43
watched a 3-minute instructional video that told them to select two images that remind them of
personal experiences of rejection/abandonment and three personalized abandonment phrases that
described each of the two images.
Participants who completed Part I signed up for a Part II time-slot that occurred
approximately 1 week later. Part II occurred on campus and involved two tasks. First, to assess
secure base script knowledge, the narrative-based Adult Script Assessment (ASA) was
administered by a trained undergraduate research assistant (RA) using four word-prompt cards in
which participants told stories with attachment-related themes from a set of words. Second,
participants were taken through the Abandonment Stroop Task (AST).
The task began with 16 practice trials. For the practice trials, the primes were phrases
describing categories (e.g., A KIND OF TREE, A CLOSE RELATIVE) superimposed on a white
slide and randomized to appear in one of the four corners of the slide for each trial, and the
targets were category members (e.g., OAK, AUNT). On half the trials, the target was a member of
the category presented in the prime (e.g., A KIND OF FAST FOOD, FRIES), and in the
remaining half the category of the prime and target did not match (e.g., A KIND OF WEAPON
and PIANO). The 4 primes and 8 targets were randomly ordered with the constraint that the
primes appeared once with a same-category target and once with a different-category target,
giving 8 prime-target pairs. This ordering of prime-target pairs was repeated, yielding 16 total
practice trials. The RA completed the first 4 trials to demonstrate how the task works to the
participant. The participant completed the remaining 12 practice trials and was given an
opportunity to ask any remaining questions prior to beginning the experimental trials.
Participants then completed the 96 experimental trials and the RA used a form to record any
errors made in stating the target word, prime phrase, or instances in which the E-Prime Software
did not register the participant’s voice. Finally, participants were thanked for their participation
were awarded 1 credit for participation or paid $15 depending on how they were recruited.
Measures
Attachment Script Assessment (ASA) – To assess secure base script knowledge the narrative-
based ASA was administered using four word-prompt cards in which participants tell stories
with attachment-related themes from a set of words. Each card had a title for each story and a list
of 12 words (subdivided into three columns). Participants were told to tell the best story they can
tell using the words as an outline and were instructed to speak for approximately 2-3 minutes for

44
each of four cards (two adult-child narratives and two adult-adult narratives). Narratives were
transcribed and coded on a 7-point scale (1 = no secure base script with atypical script content; 7
= rich and detailed secure base script) for the extent to which they demonstrated secure base
knowledge. The ASA has demonstrated strong reliability, validity in terms of consistent
associations with attachment interview data, and cross-time stability (Vaughn et al., 2006).
Abandonment Stroop Task (AST) – To assess the degree to which rejected self-states and
abandoning other meaning elements are organized in a cognitive representation among those
with low secure base knowledge, an AST was constructed based on similar Stroop tasks
(Dandeneau & Baldwin, 2004; Segal et al., 1995) (see design above). The AST was built using
E-Prime 2.0 software and voice response was captured using the Chronos USB-based response
device.
Results
Reaction times (RTs) for trials in which the participants made errors were not included in
the analysis. The number of errors made was not associated with any of the other variables
examined in this study. Following Milanak and Berenbaum (2014), reaction times shorter than
300 milliseconds were eliminated. In addition, RTs were removed from analysis if they were: (1)
smaller than and/or larger than 3 standard deviations from the group grand mean and (2) smaller
than and/or larger than 3 standard deviations from the individual participant’s mean. To correct
for non-normality of the prime and target response latencies, we used a reciprocal (1/RT)
transformation (Ratcliff, 1993).
We hypothesized that there would be a three-way interaction between secure-base script
score (ASA) and latency in naming of word by prime slide and target word. In order to test this
prediction, we conducted a repeated measures ANOVA entering Prime as the first factor with
two levels (NN, NA), Target as the second factor with two levels (AA, AN), and ASA as a
covariate. We then tested this three-way (ASA x Prime x Target) interaction to see whether there
was a significant ASA x Prime x Target effect for response latency.

Table 11. Summary of Repeated Measures Analysis of Variance for Latency in Response Time
Predicted by Prime, Target, and the Adult Script Assessment (ASA)
2
Effect MS df F p Partial η
Prime 1.08*10^-10 1 0.02 0.90 0.00
Prime x ASA 8.27*10^-11 1 0.01 0.91 0.00
Error 6.23*10^-9 207

45
Table 11 (continued).

Target 1.01*10^-8 1 1.70 0.19 0.01

Target x ASA 1.03*10^-8 1 1.74 0.19 0.01
Error 5.92*10^-9 207
Prime x Target 1.06*10^-8 1 1.38 0.24 0.01
Prime x Target x ASA 1.19*10^-8 1 1.56 0.21 0.01
Error 7.67*10^-9 207

Note. Prime = Neutral or Abandonment, Target = Neutral or Abandonment, and ASA = Adult Script
Assessment. * p < .05.

As can be seen in Table 11 above, neither the main effects nor the interactions were statistically
significant. Effect sizes were small for Target, Target x ASA, and Prime x Target x ASA. In
summary, contrary to our prediction, a significant three-way interaction (Prime x Target x ASA)
was not observed.
Discussion
The present study aimed to test whether the organization of representational content of an
abandonment schema could be elicited through an emotional Stroop task. Our Abandonment
Stroop Task (AST) involved presentation of personalized prime slides that depicted phrases and
images related to each participant’s past experiences of abandonment. If color naming of
abandonment target words was slower when primed by abandonment primes than by neutral
primes, it would suggest that the specific interference observed in individuals with a lack of
consolidated secure base script is mediated by a representation in which this information is
organized in this manner. We hypothesized that individuals with low secure base script
knowledge would show the longest color-naming latencies when an abandoning self-relevant
prime slide was paired with an abandonment target word. However, a statistically significant
three-way interaction (Prime x Target x ASA) was not observed. This finding is curious given
that prior emotional Stroop tasks similar to our design have found this kind of self-other
organization among depressed individuals (Segal, Gemar, Truchon, Guirguis, & Horowitz,
1995).
One possible explanation for our results is that our recruited sample lacked a sufficient
number of individuals with low secure base script knowledge to detect the effect. In particular,
we struggled to recruit participants low on the ASA which led to a positive skew in ASA scores
in our sample (see Figure 8 below).

46
Figure 8. A histogram of Adult Script Assessment (ASA) scores for participants in Study 2 (N = 209).

Nonetheless, as can be seen in Figure 8, we did have a substantial number of participants with
high levels of secure base script knowledge (i.e., ASA scores > 4). In addition, we did surpass
our target sample size of 200 based on an a priori power analysis.
In this context, the most direct explanation for the failure to find a significant three-way
interaction is that the representational content of an abandonment schema is not organized in the
form we hypothesized. That is, contrary to claims in the internal working models literature, self-
other meaning elements associated with abandonment may not be organized such that the
priming of one element activates the other elements. It may be that significant experiences of
abandonment distress are not linked directly to semantic knowledge regarding the specific details
of events like those we primed in this study. Alternatively, priming may trigger a shift in
phenomenological perspective that would be better detected by a projective measure such as the
SCORS rather than interference in cognitive control as assessed in this study.
Potential limitations of the present study include use of a non-clinical sample, which may
have reduced our ability to effectively prime participants’ past abandonment experiences.
Indeed, prior studies using the current Stroop paradigm have focused on comparing clinical to
non-clinical populations (e.g. depressed versus non-depressed individuals) that may be more
likely to exhibit priming effects associated with schema activation. Second, although we

47
developed a means of personalizing the prime slides based on participants’ past abandonment
narratives, the prime slide pictures and phrases likely had a high degree of variability in intensity
across participants. Nonetheless, the AST we designed and our hypothesizing of a three-way
interaction had a strong theoretical rationale. In addition, our recruitment of a large sample based
on a prior power analysis reduced the likelihood of a false positive result.

48
 CHAPTER 4: REFLECTIONS ON A NEW MECHANISTIC APPROACH TO
RELATIONAL SCHEMAS
Memory is the process by which past experience informs present experience, thereby
guiding human behavior. In this vein, attachment theorists have argued that “internal working
models” may capture the way early learning history is internalized to facilitate response to
interpersonal threats later in life. Although an extensive literature has demonstrated that early
attachment experiences indeed inform adult behavior, limitations to the attachment approach
include: (1) lack of clarity regarding the content of internal working models, (2) limited
explanation of how working models function to direct behavior in stressful or rejecting
situations, and (3) difficulty reconciling the categorical nature of attachment designations with
the cross-temporal instability of internal working models found in longitudinal studies.
The current research aimed to improve upon the attachment approach by providing a
more specific conceptualization of an internal working model as a cognitive schema. Based on
learning theory, the schema was conceptualized as being composed of an implicit expectation of
being abandoned, implicit memories of abandonment distress, and low secure base script
knowledge. The empirical studies described in Chapters 2 and 3 examined whether this specific
memory mechanism, a relational schema of abandonment, could be measured and activated such
that its causal influence on experience and behavior could be elucidated. Although the findings
in the two studies were different than expected, several conclusions can be drawn from this
research that have relevance for understanding the etiology of psychopathology and for
therapeutic intervention.
Study 1 demonstrated that varying the meaning of a statement in a socially charged
situation elicited different implicit emotional reactivity depending on whether individuals were
high or low on secure base script knowledge. This difference could not be accounted for by any
explicit measure of experience or behavior. The finding supports the contention that implicit
processes associated with schema activation at times may be more effectively assessed through
projective and psychophysiological measures than through self-report. In other words, the
experiment illustrates how bodily reactivity can be dissociated from explicit knowledge. On the
other hand, explicit measures demonstrated better predictive utility for behaviorally observed
distress. The YSQ may be well suited to capture behavioral patterns that are self-observed and
amenable to articulation in language (e.g., “I tend to get clingy when relationships end”). The

49
fact that the AB, MA, and SI schemas predicted observed distress suggests that explicit and
implicit schema measures tap different aspects of memory with different behavioral
consequences.
The difference in response to an ambiguous (non-rejection) stimulus versus a rejection
stimulus added complexity to the observed outcomes. Since schemas are thought to be latent
memory mechanisms that are activated in schema-congruent contexts, we hypothesized that the
rejection condition would produce the greatest differences in outcomes based on schema status.
Instead, the AB, MA, and SI schemas were more predictive of large differences in reaction to the
non-rejection stimulus. Coupled with the findings from our measurement of skin conductance
response, it appears that situations in which the meaning of a relational event is open to
interpretation may be highly relevant to how schemas work. It may be that these are the very
situations in which “consistency” is maintained and memory supplies an interpretation of reality
that differs significantly between those high and low on the schema.
Finally, Study 1 provided an initial step towards understanding the cross-temporal
instability of attachment. Single time-point designs that draw conclusions from correlations
between measures of attachment, schemas, and signs and symptoms of psychopathology are ill
equipped to make any causal inferences regarding the dynamic connection between schema
activation, experience, and behavior. By contrast, Study 1 was able to capture what kinds of
shifts in phenomenological perspective occur immediately following an ambiguous (non-
rejection) and rejection stimulus. The ASA identified participants low in secure base script who
self-reported as having a secure relationship style, and theses individuals demonstrated less of an
increase in positive expectations of others and more loss of control over anger following the
rejection manipulation. All other participants had this reaction to the non-rejection manipulation.
This raises the possibility that these “low insight” individuals are at higher risk for
psychopathology because they lack the internal resources (including secure base script
knowledge) necessary to deal with their emotional reactions to rejection. Perhaps through a
psychological intervention these individuals could learn to observe their reactions and develop
the skills to self-soothe. This shift in self-awareness might paradoxically lead them to self-report
as non-secure precisely when they are able to manage their problematic reactions to attachment-
related situations.

50
 Although a variety of generative conclusions could be drawn from Study 1, the results of
Study 2 proved to be more valuable in terms of what it appeared to refute. The most likely
conclusion from Study 2 is that content associated with abandonment distress is not organized in
terms of discrete self-other meaning elements. The lack of effect of prime and target in the
Stroop task suggests that even if abandonment themes were elicited in participants, this
momentary elicitation did not significantly interfere with cognitive performance.
It should be noted that the memory mechanism in Figure 1 was a theoretical conjecture
that relied heavily on the ASA as a measure that would be reliably related to abandonment
distress. The measure did appear useful in demonstrating how bodily memory can be dissociated
from explicit self-knowledge. It was also effective as an index of differential response to an
ambiguous social stimulus. Nonetheless, whether secure base script is an essential measure for
understanding the role of an abandonment schema in psychopathology remains an open question.
As it is likely that a number of other measures could better elucidate these phenomena, it is
important to return to the phenomena being investigated. A program of research based on the
empirical studies presented here would not limit itself to the ASA, but would be focused on
additional ways to experimentally assess the dynamics of schema activation.
An important factor to consider in assessing these dynamics in future work is the wide
variety of events that can be experienced as abandonment and the differential implications of
these events at different ages. In addition to experiences with primary caretakers early on in life,
it is crucial to bring attention to the variety of ways individuals may experience abandonment
through peer groups, significant romantic relationships, death of a loved one, or a range of other
circumstances. The developmental consequences of these events undoubtedly vary based on
when they occur in the lifespan. Take for instance an individual who grows up in a home with
relatively supportive caretakers who nurture the capacity for self-soothing and attachment
security. If this individual loses a best friend in combat at age 20, how does this individual’s
attempt to cope with abandonment differ than someone who experienced emotional abuse as a
child? The implications of this question involve both the cumulative effect of negative life
experiences and the role of critical periods in the development of relational schemas.
The highly contextual and idiographic nature of life experience and memory makes
clarifying these issues quite challenging. The most important implication is the need to resist
thinking of abandonment as a monolithic phenomenon. Further investigation might attend to

51
ways that individuals describe experiences of abandonment. In terms of schema development,
are there patterns to the phenomenology of abandonment that point to what needs, desires, hopes,
and/or expectations these experiences tend to thwart? In terms of schema activation, how is the
intensity of the experience, its chronicity, and the age at which it occurs related to how it is
instantiated in memory and what later events trigger associated behavioral patterns? In terms of
schema-related behavior, how does the activation of information structures influence
subjectively perceived emotional states and coping responses?
These questions overlap significantly with the concept of affective chronometry—the
temporal dynamics of affective responding (Davidson, 2015). In particular, a potentially fruitful
avenue for future research might include examining individual differences in the length of time it
takes for the physiological response to social threat to subside. The duration of subjectively
experienced emotion might be tracked at the same time. It would be interesting to see if a
relationship between these variables has bearing on the likelihood an individual will exhibit
psychopathological signs and symptoms. This methodology could be used to probe whether
awareness of one’s physiological state is a hallmark of healthy adaptation or whether it indicates
potential anxiety-inducing interoceptive monitoring. Perhaps dysfunctional outcomes are most
likely when physiological arousal persists with limited subjective experiencing. Distress might
then be manifested in compensatory behaviors (e.g., over-eating, anger outbursts) that could be
observed on longer time scales (e.g., through event sampling, daily diaries, or ecological
momentary assessment).
Future work might also more richly examine how re-experiencing abandonment distress
may elicit shifts in phenomenological perspective and/or longer lasting changes in mood. Part of
the challenge may be to understand how more generalized instantiations of mood or bodily
memories are stored and retrieved, and how they then inform moment-to-moment experience and
behavior. It may be that mixed methods research is needed that can track how the
phenomenology of schema activation and behavior co-vary. Of interest would be “thick
descriptions” of what the re-experiencing of distress is like, combined with observational data
(e.g., facial expression, coding of non-verbal behavior), and psychophysiological measures (e.g.,
heart rate, skin conductance, muscle tension, etc.) of moment-to-moment behavioral responses.
Overall, the current research outlines a new mechanistic approach to understanding
relational schemas and their involvement in psychological distress. The novelty of the empirical

52
studies described here rests on the attempt to examine the dynamic effects of schema activation
in meaningful social situations. The broad conclusion that schemas may have particular
relevance in ambiguous environments (such that the individual imposes a meaning to maintain
intelligibility) warrants future experimental corroboration. Studies with this aim must elicit
schematic behavior in situations in which the individual is unaware of the true experimental aims
and observations. Additionally, they will make use of measures (e.g., projective and
psychophysiological) that tell us how a person is responding without relying solely on self-
report. These measures need not fit within a narrow conception of “projective” or “implicit.” For
instance, perhaps a social interaction task in an online “chat-room” could be used to assess
patterns in relational responding to others. Here the data obtained may reflect tendencies of an
individual that are outside of awareness but are nonetheless displayed in concrete ways that
require less steps of interpretation than projective measures that utilize open-ended responses to
prompt words/pictures. It is possible to survey participants’ beliefs about their own behavior
prior to contrasting this with this kind of covert observation. The overarching perspective that
can frame this kind of future schema research is the idea that a range of types of data (e.g.,
qualitative, implicit, explicit, physiological) is necessary to arrive at a holistic understanding of
the role of memory structures in the moment-to-moment manifestation of psychological distress.

53
 REFERENCES
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. N. (2015). Patterns of attachment: A
psychological study of the strange situation. Psychology Press.
Bach, B., Lee, C., Mortensen, E. L., & Simonsen, E. (2015). How do DSM-5 personality traits
align with schema therapy constructs? 29, 212.
Baldwin, M. W. (1992). Relational schemas and the processing of social information.
Psychological Bulletin, 112(3), 461–484.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/0033-2909.112.3.461
Baldwin, M. W., & Sinclair, L. (1996). Self-esteem and “if…then” contingencies of
interpersonal acceptance. Journal of Personality and Social Psychology, 71(6), 1130–
1141. http://dx.doi.org/10.1037/0022-3514.71.6.1130
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a
four-category model. Journal of Personality and Social Psychology, 61(2), 226–244.
http://dx.doi.org/10.1037/0022-3514.61.2.226
Bowlby, J. (1969). Attachment and loss: Vol. I. Attachment. New York: Basic Books.
Calvete, E., Estévez, A., López de Arroyabe, E., & Ruiz, P. (2005). The Schema Questionnaire--
Short Form: Structure and Relationship with Automatic Thoughts and Symptoms of
Affective Disorders. European Journal of Psychological Assessment, 21(2), 90–99.
http://dx.doi.org.proxy2.library.illinois.edu/10.1027/1015-5759.21.2.90
Coppola, G., Vaughn, B. E., Cassibba, R., & Costantini, A. (2006). The attachment script
representation procedure in an Italian sample: Associations with adult attachment
Interview scales and with maternal sensitivity. Attachment & Human Development, 8(3),
209–219. https://doi.org/10.1080/14616730600856065
Dandeneau, S. D., & Baldwin, M. W. (2004). The Inhibition of Socially Rejecting Information
Among People with High Versus Low Self-Esteem: The Role of Attentional Bias and the
Effects of Bias Reduction Training. Journal of Social and Clinical Psychology, 23(4),
584–603. https://doi.org/10.1521/jscp.23.4.584.40306
Davidson, R. J. (2015). Comment: Affective Chronometry Has Come of Age. Emotion Review,
7(4), 368–370. https://doi.org/10.1177/1754073915590844

54
Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate
relationships. Journal of Personality and Social Psychology, 70(6), 1327–1343.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/0022-3514.70.6.1327
Dozier, M., Stovall-McClough, K. C., & Albus, K. E. (2008). Attachment and psychopathology
in adulthood. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory,
research, and clinical applications (pp. 718–744). New York, NY, US: Guilford Press.
Dykas, M. J., Woodhouse, S. S., Cassidy, J., & Waters, H. S. (2006). Narrative assessment of
attachment representations: Links between secure base scripts and adolescent attachment.
Attachment & Human Development, 8(3), 221–240.
https://doi.org/10.1080/14616730600856099
Ehrhart, M. G., Ehrhart, K. H., Roesch, S. C., Chung-Herrera, B. G., Nadler, K., & Bradshaw, K.
(2009). Testing the latent factor structure and construct validity of the Ten-Item
Personality Inventory. Personality and Individual Differences, 47(8), 900–905.
https://doi.org/10.1016/j.paid.2009.07.012
Fraley, R. C., Roisman, G. I., & Haltigan, J. D. (2013). The legacy of early experiences in
development: Formalizing alternative models of how early experiences are carried
forward over time. Developmental Psychology, 49(1), 109–126.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/a0027852
Hibbard, S., Hilsenroth, M. J., Hibbard, J. K., & Nash, M. R. (1995). A validity study of two
projective object representations measures. Psychological Assessment, 7(4), 432–439.
http://dx.doi.org/10.1037/1040-3590.7.4.432
Hinde, R. (1988). Continuities and discontinuities. In M. Rutter (Ed.), Studies of psychosocial
risk: The power of longitudinal data (pp. 367–384). Cambridge University Press.
Hoffart, A., Sexton, H., Hedley, L. M., Wang, C. E., Holthe, H., Haugum, J. A., … Holte, A.
(2005). The Structure of Maladaptive Schemas: A Confirmatory Factor Analysis and a
Psychometric Evaluation of Factor-Derived Scales. Cognitive Therapy and Research,
29(6), 627–644. https://doi.org/10.1007/s10608-005-9630-0
Hur, J., Iordan, A. D., Berenbaum, H., & Dolcos, F. (2016). Emotion–attention interactions in
fear conditioning: Moderation by executive load, neuroticism, and awareness. Biological
Psychology, 121, 213–220. https://doi.org/10.1016/j.biopsycho.2015.10.007

55
Iacobucci, D., Schneider, M. J., Popovich, D. L., & Bakamitsos, G. A. (2016). Mean centering
helps alleviate “micro” but not “macro” multicollinearity. Behavior Research Methods,
48(4), 1308–1317. https://doi.org/10.3758/s13428-015-0624-x
Mason, O., Platts, H., & Tyson, M. (2005). Early maladaptive schemas and adult attachment in a
UK clinical population. Psychology & Psychotherapy: Theory, Research & Practice,
78(4), 549–564. https://doi.org/10.1348/147608305X41371
McLean, H. R., Bailey, H. N., & Lumley, M. N. (2014). The secure base script: Associated with
early maladaptive schemas related to attachment. Psychology & Psychotherapy: Theory,
Research & Practice, 87(4), 425–446. https://doi.org/10.1111/papt.12025
Mikulincer, M., Shaver, P. R., Sapir-Lavid, Y., & Avihou-Kanza, N. (2009). What’s inside the
minds of securely and insecurely attached people? The secure-base script and its
associations with attachment-style dimensions. Journal of Personality and Social
Psychology: Interpersonal Relations and Group Processes, 97(4), 615–633.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/a0015649
Muran, J. C. (1991). A reformulation of the ABC model in cognitive psychotherapies:
Implications for assessment and treatment. Clinical Psychology Review, 11(4), 399–418.
https://doi.org/10.1016/0272-7358(91)90115-B
Muran, J., Samstag, L., Segal, Z., & Winston, A. (1998). Interpersonal Scenarios: An Idiographic
Measure of Self-Schemas. Psychotherapy Research, 8(3), 321–333.
https://doi.org/10.1080/10503309812331332417
Nelson, K. (1999). Event representations, narrative development and internal working models.
Attachment & Human Development, 1(3), 239–252.
https://doi.org/10.1080/14616739900134131
Peters, E. J., Hilsenroth, M. J., Eudell-Simmons, E. M., Blagys, M. D., & Handler, L. (2006).
Reliability and validity of the Social Cognition and Object Relations Scale in clinical use.
Psychotherapy Research, 16(5), 617–626. https://doi.org/10.1080/10503300600591288
Pietromonaco, P. R., & Barrett, L. F. (2000). The internal working models concept: What do we
really know about the self in relation to others? Review of General Psychology, 4(2),
155–175. http://dx.doi.org.proxy2.library.illinois.edu/10.1037/1089-2680.4.2.155
Rijkeboer, M. M., van den Bergh, H., & van den Bout, J. (2005). Stability and discriminative
power of the Young Schema-Questionnaire in a Dutch clinical versus non-clinical

56
 population. Journal of Behavior Therapy and Experimental Psychiatry, 36(2), 129–144.
https://doi.org/10.1016/j.jbtep.2004.08.005
Rosebrock, L. E., Hoxha, D., Norris, C., Cacioppo, J. T., & Gollan, J. K. (2017). Skin
conductance and subjective arousal in anxiety, depression, and comorbidity: Implications
for affective reactivity. Journal of Psychophysiology, 31(4), 145–157.
http://dx.doi.org.proxy2.library.illinois.edu/10.1027/0269-8803/a000176
Schank, R. C., & Abelson, R. P. (2013). Scripts, plans, goals, and understanding: An inquiry
into human knowledge structures. Psychology Press.
Schmidt, N. B., Jr, T. E. J., Young, J. E., & Telch, M. J. (1995). The schema questionnaire:
Investigation of psychometric properties and the hierarchical structure of a measure of
maladaptive schemas. Cognitive Therapy and Research, 19(3), 295–321.
https://doi.org/10.1007/BF02230402
Segal, Z. V. (1988). Appraisal of the self-schema construct in cognitive models of depression.
Psychological Bulletin, 103(2), 147–162.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/0033-2909.103.2.147
Segal, Z. V., Gemar, M., Truchon, C., Guirguis, M., & Horowitz, L. M. (1995). A priming
methodology for studying self-representation in major depressive disorder. Journal of
Abnormal Psychology, 104(1), 205–213. http://dx.doi.org/10.1037/0021-843X.104.1.205
Segal, Z. V., & Shaw, B. F. (1986). Cognition in depression: A reappraisal of Coyne and
Gotlib’s critique. Cognitive Therapy and Research, 10(6), 671–693.
https://doi.org/10.1007/BF01173754
Steele, R. D., Waters, T. E. A., Bost, K. K., Vaughn, B. E., Truitt, W., Waters, H. S., …
Roisman, G. I. (2014). Caregiving antecedents of secure base script knowledge: A
comparative analysis of young adult attachment representations. Developmental
Psychology, 50(11), 2526–2538.
http://dx.doi.org.proxy2.library.illinois.edu/10.1037/a0037992
Thompson, R. A. (2008). Attachment-related mental representations: Introduction to the special
issue. Attachment & Human Development, 10(4), 347–358.
https://doi.org/10.1080/14616730802461334

57
Thompson, R. A., & Raikes, H. A. (2003). Toward the next quarter-century: Conceptual and
methodological challenges for attachment theory. Development and Psychopathology,
15(3), 691–718. https://doi.org/10.1017/S0954579403000348
Van der Kolk, B. A., McFarlane, A. C., & Weisaeth, L. (1996). Traumatic stress. New York,
NY, US: Guilford Press.
Van Vlierberghe, L., Braet, C., Bosmans, G., Rosseel, Y., & Bögels, S. (2010). Maladaptive
Schemas and Psychopathology in Adolescence: On the Utility of Young’s Schema
Theory in Youth. Cognitive Therapy & Research, 34(4), 316–332.
https://doi.org/10.1007/s10608-009-9283-5
Vaughn, B. E., Veríssimo, M., Coppola, G., Bost, K. K., Shin, N., McBride, B., … Korth, B.
(2006). Maternal attachment script representations: Longitudinal stability and
associations with stylistic features of maternal narratives. Attachment & Human
Development, 8(3), 199–208. https://doi.org/10.1080/14616730600856024
Waters, H. S., & Waters, E. (2006). The attachment working models concept: Among other
things, we build script-like representations of secure base experiences. Attachment &
Human Development, 8(3), 185–197. https://doi.org/10.1080/14616730600856016
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: The PANAS scales. Journal of Personality and Social
Psychology, 54(6), 1063–1070. http://dx.doi.org/10.1037/0022-3514.54.6.1063
Welburn, K., Coristine, M., Dagg, P., Pontefract, A., & Jordan, S. (2002). The Schema
Questionnaire—Short Form: Factor Analysis and Relationship Between Schemas and
Symptoms. Cognitive Therapy & Research, 26(4), 519.
Young, J. E. (2003). Schema therapy: A practitioner’s guide /. New York : Guilford Press.

58
APPENDIX A – IRB LETTERS & DEBRIEFINGS

59
 Research Participant Debriefing Form
Department of Psychology
University of Illinois at Urbana-Champaign

Thank you for participation in our study. Your time and effort are greatly appreciated!

Although you were told that the study was about biological markers and social factors that
influence the way people construct narratives, the true purpose of this study was to explore
whether or not a mental representation (schema) of abandonment can be measured, activated,
and linked to specific emotional and behavioral responses following interpersonal rejection. A
schema can be defined as an unconscious mental map that guides how a person interprets and
responds to information in the environment. A number of schemas are thought to form early in
life based on experiences that an individual has with caretakers. It is hypothesized that
individuals who do not receive sufficient care and nurturance in childhood may develop an
abandonment schema that is adaptive at the time, but later becomes maladaptive in new
situations as an adult.

In order to test this hypothesis, we needed to create a social interaction task where you believed
that another participant was completing the same study. Your social interaction task was in fact
with a lab member who was acting as a confederate (a person who pretends they are a
participant). This deception was necessary in order to test the hypothesis that some individuals
may respond with more emotional intensity than others following perceived rejection due to an
abandonment schema.

We wish to be clear about the nature of this deception. The lab member who you interacted with
did NOT know what you would be told following the first interaction. The message the lab RA
told you therefore had nothing to do with how you carried yourself or responded in the social
interaction task. You will now have a chance to meet the lab member and we will address any
remaining questions you might have. Below are a few pieces of literature on schemas and
rejection that you can consult if you are interested in learning more about the basis of this study.
In addition, we are happy to discuss with you further why we think this study adds to our
understanding of psychological distress.

• Chapter 1 – Schema Therapy: Conceptual Model

Young, J. E. (2003). Schema therapy : A practitioner’s guide. New York : Guilford Press.

• Chapter 1 – The Interpersonal Process Approach

Teyber, E., & McClure, F. (2010). Interpersonal Process in Therapy: An Integrative
Model. Cengage Learning.
• Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate
relationships. Journal of Personality and Social Psychology, 70(6), 1327–1343.

Any questions about this study can be directed to Dr. Howard Berenbaum at
hberenba@illnois.edu or student investigator Joel Thomas at jthoma11@illinois.edu.

60
61
 Research Participant Debriefing Form

Department of Psychology
University of Illinois at Urbana-Champaign

Thank you for participation in our study. Your time and effort are greatly appreciated!

The purpose of this study was to explore whether or not a mental representation (schema) of
abandonment can be measured, and whether this schema contains stored memories of personal
events. A schema can be defined as an unconscious mental map that guides how a person
interprets and responds to information in the environment. A number of schemas are thought to
form early in life based on experiences that an individual has with caretakers. It is hypothesized
that individuals who do not receive sufficient care and nurturance in childhood may develop an
abandonment schema that is adaptive at the time, but later becomes maladaptive in new
situations as an adult.

The task you completed today first assessed whether phrases and words related to abandonment
are interconnected in your memory system. The personal picture and phrase further assessed
whether memories of your personal experiences made it more difficult for you to complete the
task. Below are a few pieces of literature on schemas and rejection that you can consult if you
are interested in learning more about the basis of this study. In addition, we are happy to discuss
with you further why we think this study adds to our understanding of psychological distress.

• Chapter 1 – The Interpersonal Process Approach

Teyber, E., & McClure, F. (2010). Interpersonal Process in Therapy: An Integrative
Model. Cengage Learning.

• Downey, G. & Feldman, S. I. (1996). Implications of Rejection Sensitivity for Intimate

Relationships. Journal of Personality and Social Psychology, 70(6) 1327-1343.

• Dandeneau, S. D., & Baldwin, M. W. (2004). The inhibition of socially rejecting

information among people with high versus low self-esteem: The role of attentional bias
and the effects of bias reduction training. Journal of Social and Clinical Psychology,
23(4), 584-602.

Any questions about this study can be directed to Dr. Howard Berenbaum at
hberenba@illnois.edu or student investigator Joel Thomas at jthoma11@illinois.edu

62
 APPENDIX B – MEASURES
Demographic Questionnaire
Instructions: Please type in or check the following information about yourself. All
responses are confidential. Please answer the questions as honestly and thoroughly as
possible. While we ask that you fill out all questions, you can choose to skip anything
that you do not want to answer.

Background Information
1. How old are you? __________________ years
2. Gender:

Female Male

Transgender Genderqueer

Other

3. What is your year in college?

Freshman Sophomore

Junior Senior

Other

4. How long have you lived in the US?

< 1year 1-2 years 2-3 years 3-4 years

4-5 years 5-7 years 7-9 years 9-12 years

12-15 years 15-20 years > 20 years

5. What is your ethnic background?

Hispanic or Latina/Latino Not Hispanic or Latina/Latino

Unknown or prefer not to report

6. How would you describe your racial background? (Please check all that apply)

Caucasian/White Pacific Islander or Hawaiian

Black or African American Native American or Alaska Native

Asian or Asian American Indian Subcontinent

63
 Other (specify): _______________

Mixed racial/ethnic heritage (specify): _______

7. What is your relationship status? (Please check all that apply)

Single, never married Married/ Civil Union

Casually Dating Separated

Long-Term Relationship, not living together Divorced

Living with someone Widowed

Other________________________________________________________

8. What is your preferred romantic partner gender?

Female Male

Transgender Genderqueer

Other

Language:
9. Please rate your overall ability in the English language:

1 2 3 4
Some familiarity Limited Competent in Native English
competence (e.g., speaking, reading, speaker
speaking but not & writing
writing)

10. AFTER you have completed your experiments for course credit, would you be
interested in participating in other psychological experiments that you would be paid
for?
Yes
No

64
 Young Schema Questionnaire – short version (2005)
(YSQ – S3)

Name Date

INSTRUCTIONS
Listed below are statements that people might use to describe themselves.
Please read each statement, then rate it based on how accurately it fits you
over the past year. When you are not sure, base your answer on what you
emotionally feel, not on what you think to be true.

A few of the items ask about your relationships with your parents or romantic
partners. If any of these people have died, please answer these items based on
your relationships while they were alive. If you do not currently have a partner
but had partners in the past, please answer the item based on your most recent
significant romantic partner.

Choose the highest score from 1 to 6 on the rating scale below that best
describes you, then write your answer in the block on the right of each
statement.

RATING SCALE
1 = Completely untrue of me 4 = Moderately true of me
2 = Mostly untrue of me 5 = Mostly true of me
3 = Slightly more true than untrue 6 = Describes me perfectly

1 I haven't had someone to nurture me, share him/herself with me,

or care deeply about everything that happens to me.
2 I find myself clinging to people I'm close to because I am afraid
they’ll leave me.

3 I feel that people will take advantage of me.

4 I don't fit in.

65
5 No man/woman I desire could love me once he or she saw my
defects.

6 I don’t have people to give me warmth, holding and affection

7 I need other people so much that I worry about losing them

I feel that I cannot let my guard down in the presence of other

8 people, or else they will intentionally hurt me

9 I'm fundamentally different from other people

10 No one I desire would want to stay close to me if he or she knew

the real me

11 I haven't felt that I am special to someone

12 I worry that people I feel close to will leave me or abandon me

13 It is only a matter of time before someone betrays me

14 I don't belong; I'm a loner

15 I'm unworthy of the love, attention and respect of others

I have not had someone who really listens to me, understands
16 me or is tuned into my true needs and feelings

When someone I care for seems to be pulling away or

17 withdrawing from me, I feel desperate

18 I am quite suspicious of other people's motives

19 I feel alienated or cut off from other people

66
 20 I feel that I’m not lovable
I haven’t had a strong or wise person to give me sound advice
21 or direction when I am not sure what to do

22 Sometimes I am so worried about people leaving me that I drive

them away

23 I am usually on the lookout for other people’s ulterior or hidden

motives

24 I always feel on the outside of groups

I am too unacceptable in very basic ways to reveal myself to

25 other people or let them get to know me well

© 2005 Jeffrey Young, Ph.D. Special thanks to Gary Brown, Ph.D., Scott Kellogg,
Ph.D., Glenn Waller,, Ph.D., and the many other therapists and researchers who
contributed items and feedback in the development of the YSQ. Unauthorized
reproduction, translation or modification in any form whatsoever without written
consent of the author is prohibited. For more information write: Schema Therapy
Institute, 36 West 44th Street, Suite. 1007, New York, NY 10036.

Prompt-word outlines (attachment script assessment).

A. Baby’s Morning (mother – child attachment story)

mother hug teddy bear

baby smile lost
play story found
blanket pretend nap

B. Doctor’s Office (mother – child attachment story)

Charlie hurry mother

bike doctor toy
hurt cry stop
mother shot hold

C. Taylor and Chase’s Camping Trip (adult – adult attachment story)

67
Taylor tent campfire
Chase wind shadow
bags collapse sounds
hurry upset hug

D. Sam’s Accident (adult – adult attachment story)

Sam wait home

road Alex dinner
accident tears bed
hospital doctor hug

Interaction Questionnaire

Overall, how well do you feel the first interaction period went?

1 2 3 4 5
Very poorly Poorly Neutral Fairly well Very well

Are you looking forward to meeting the other person again?

YES
NO
INDIFFERENT

68
STATE PANAS FOR NEGATIVE AFFECT - This scale consists of a number of words and
phrases that describe different feelings and emotions. Read each item and then mark the
appropriate answer on the corresponding circle on your scantron sheet. Indicate to what extent
you feel this way AT THIS MOMENT. Use these letters to record your answers.

A B C D E
Very slightly A little Moderately Quite a bit Extremely
or not at all

1. guilty
2. afraid

3. nervous

4. excited
5. unaccepted

6. hostile

7. jittery
8. upset

9. inspired
10. rejected

11. enthusiastic

12. distressed
13. determined

14. proud
15. hurt

16. interested

17. active
18. attentive

19. strong

20. disliked
21. alert

69
 22. irritable

23. ashamed
24. scared

25. discouraged
26. frustrated

27. down

28. sad
29. happy

30. satisfied

31. tranquil

Observed Emotional Reaction Form

Which of the following is most descriptive of the participant’s facial
expression post-manipulation?

1 2 3 4
Not upset Slightly upset Upset Very upset

70
TAT SOCIAL COGNITIONS AND OBJECT RELATIONS TASK PICTURES

71
72
73
74
75
76
 THE RELATIONSHIPS QUESTIONNAIRE (RQ)
Scale:
Following are four general relationship styles that people often report. Place a checkmark
next to the letter corresponding to the style that best describes you or is closest to the way
you are.

secure A. It is easy for me to become emotionally close to others. I am

comfortable depending on them and having them depend on me. I don’t worry
about being alone or having others not accept me.

fearful B. I am uncomfortable getting close to others. I want emotionally

close relationships, but I find it difficult to trust others completely, or to depend
on them. I worry that I will be hurt if I allow myself to become too close to
others.

preoccupied C. I want to be completely emotionally intimate with others, but I

often find that others are reluctant to get as close as I would like. I am
uncomfortable being without close relationships, but I sometimes worry that others
don’t value me as much as I value them.

dismissing D. I am comfortable without close emotional relationships. It

is very important to me to feel independent and self-sufficient, and I prefer not
to depend on others or have others depend on me.
Now please rate each of the relationship styles above to indicate how well or
poorly each description corresponds to your general relationship style.

77
 Style A

1 2 3 4 5 6 7
Disagree Neutral/ Agree
Strongly Mixed Strongly

Style B

1 2 3 4 5 6 7
Disagree Neutral/ Agree
Strongly Mixed Strongly

Style C

1 2 3 4 5 6 7
Disagree Neutral/ Agree
Strongly Mixed Strongly

Style D

1 2 3 4 5 6 7
Disagree Neutral/ Agree
Strongly Mixed Strongly

78
TRAIT PANAS FOR NEGATIVE AFFECT

This scale consists of a number of words and phrases that describe different feelings and
emotions. Read each item and then mark the appropriate answer on the corresponding circle on
your scantron sheet. Indicate to what extent you have felt this way during the PAST FEW
WEEKS. Use these letters to record your answers.

A B C D E
Very slightly A little Moderately Quite a bit Extremely
or not at all

1. guilty
2. afraid

3. nervous
4. excited

5. hostile

6. jittery
7. upset

8. inspired
9. enthusiastic

10. distressed

11. determined

12. proud

13. interested

14. active
15. attentive

16. strong
17. alert

18. irritable

19. ashamed
20. scared

79
 21. frustrated

22. down
23. sad

24. happy
25. satisfied

26. tranquil

Ten-Item Personality Inventory - (TIPI)

Here are a number of personality traits that may or may not apply to you. Please write a number next
to each statement to indicate the extent to which you agree or disagree with that statement. You should
rate the extent to which the pair of traits applies to you, even if one characteristic applies more strongly
than the other.

Disagree Disagree Disagree Neither agree Agree Agree Agree

strongly moderately a little nor disagree a little moderately strongly
1 2 3 4 5 6 7
I see myself as:
1. Extraverted, enthusiastic.

2. Critical, quarrelsome.

3. Dependable, self-disciplined.

4. Anxious, easily upset.

5. Open to new experiences, complex.

6. Reserved, quiet.

7. Sympathetic, warm.

8. Disorganized, careless.

9. Calm, emotionally stable.

10. Conventional, uncreative

80
81
82
 APPENDIX C – ADDITIONAL TABLES

Table 12. Full Summary of Hierarchical Regression Analysis for Schema Variables Predicting Blind,
In-Person Coding of Observed Distress

Variable b SE b β R2 ∆R2
Step 1
Constant 2.44 0.08
Condition -0.27 0.12 -0.17*
ASA -0.03 0.06 -0.04 0.03
Step 2
Constant 2.5 0.08
Condition -0.27 0.12 -0.17*
ASA 0.04 0.08 0.04
Condition x ASA -0.14 0.12 -0.11 0.01
Step 1
Constant 2.45 0.08
Condition -0.28 0.12 -0.12*
AB schema -0.07 0.06 -0.08 0.03
Step 2
Constant 2.45 0.08
Condition -0.28 0.12 -0.17*
AB Schema -0.06 0.08 -0.08
Condition x AB Schema -0.01 0.12 -0.01 0.00
Step 1
Constant 2.45 0.08
Condition -0.28 0.12 -0.17*
MA Schema -0.07 0.06 -0.09 0.04
Step 2
Constant 2.45 0.08
Condition -0.28 0.12 -0.17*
MA Schema -0.09 0.08 -0.11
Condition x MA Schema 0.06 0.12 0.04 0.00
Step 1
Constant 2.45 0.08
Condition -0.30 0.12 -0.18*
ED Schema -0.2 0.06 -0.03 0.03
Step 2
Constant 2.45 0.08
Condition -0.30 0.12 -0.18*
ED Schema -0.04 0.08 -0.05
Condition x ED Schema 0.03 0.12 0.03 0.00
Step 1
Constant 2.44 0.08
Condition -0.27 0.12 -0.16*
SI Schema 0.06 0.06 0.07 0.03
Step 2
Constant 2.44 0.08
Condition -0.27 0.12 -0.16*

83
Table 12 (continued).

SI Schema 0.02 0.08 0.2

Condition x SI Schema 0.10 0.12 0.08 0.00
Step 1
Constant 2.42 0.08
Condition -0.27 0.12 -0.16*
DF Schema 0.08 0.06 0.10 0.04
Step 2
Constant 2.42 0.08
Condition -0.27 0.12 -0.17*
DF Schema 0.11 0.08 0.13
Condition x DF Schema -0.05 0.12 -0.04 0.00

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse Schema,
ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ Defectiveness
Schema. * p < .05.

Table 13. Full Summary of Repeated Measures Analysis of Variance for Schema Variables Predicting
Blind, Video Coding of Observed Distress

Effect MS df F p Partial
η2
Time 4.43 1 18.32 0.00 0.10*
Time x Condition 0.01 1 0.02 0.88 0.00
Time x ASA 0.01 1 0.01 0.92 0.00
Time x Cond X ASA 0.27 1 1.11 0.29 0.01
Error 0.24 167
Time 3.60 1 15.24 0.00 0.08*
Time x Condition 0.00 1 0.02 0.90 0.00
Time x AB Schema 0.32 1 1.34 0.25 0.01
Time x Cond X AB Schema 0.84 1 3.56 0.06 0.02
Error 0.24 169
Time 3.86 1 17.04 0.00 0.09*
Time x Condition 0.01 1 0.04 0.84 0.00
Time x MA Schema 1.31 1 5.80 0.02 0.03*
Time x Cond X MA Schema 2.00 1 8.87 0.00 0.05**
Error 0.23 169
Time 4.16 1 17.07 0.00 0.09*
Time x Condition 0.05 1 0.21 0.65 0.00
Time x ED Schema 0.10 1 0.40 0.53 0.00
Time x Cond X ED Schema 0.09 1 0.37 0.54 0.00
Error 0.24 167
Time 3.85 1 16.69 0.00 0.09
Time x Condition 0.01 1 0.03 0.86 0.00
Time x SI Schema 1.04 1 4.52 0.04 0.03*
Time x Cond X SI Schema 1.25 1 5.43 0.02 0.03*
Error 0.23 170
Time 3.60 1 14.72 0.00 0.08*

84
Table 13 (continued).

Time x Condition 0.03 1 0.14 0.71 0.00

Time x DF Schema 0.00 1 0.01 0.91 0.00
Time x Cond X DF Schema 0.22 1 0.89 0.35 0.01
Error 0.25 167

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse Schema,
ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ Defectiveness
Schema. * p < .05, ** p < .01.

Table 14. Full Summary of Hierarchical Regression Analysis for Condition and YSQ Schema
Variables Predicting Skin Conductance Response

Variable b SE b β R2 ∆R2
Step 1
Constant 1.03 0.19
Condition -0.91 0.27 -0.26**
AB schema -0.06 0.14 -0.04 0.07**
Step 2
Constant 1.04 0.19
Condition -0.92 0.28 -0.26**
AB Schema -0.16 0.18 -0.09
Condition x AB Schema -0.22 0.28 0.08 0.00
Step 1
Constant 1.04 0.19
Condition -0.92 0.28 -0.26**
MA Schema -0.16 0.14 -0.09 0.07**
Step 2
Constant 1.02 0.19
Condition -0.90 0.28 -0.25**
MA Schema -0.02 0.18 -0.01
Condition x MA Schema -0.33 0.28 -0.12 0.01
Step 1
Constant 1.02 0.19
Condition -0.94 0.28 -0.26**
ED Schema -0.09 0.14 -0.05 0.07**
Step 2
Constant 1.03 0.19
Condition -0.94 0.28 -0.26**
ED Schema -0.20 0.18 -0.11
Condition x ED Schema 0.28 0.29 0.10 0.01
Step 1
Constant 1.02 0.19
Condition -0.91 0.27 -0.26**
SI Schema -0.14 0.14 -0.08 0.07**
Step 2
Constant 1.03 0.19
Condition -0.91 0.28 -0.26**

85
Table 14 (continued).

SI Schema -0.19 0.19 -0.11

Condition x SI Schema 0.11 0.28 0.04 0.00
Step 1
Constant 1.03 0.20
Condition -0.91 0.28 -0.25**
DF Schema 0.08 0.14 0.05 0.06**
Step 2
Constant 1.04 0.20
Condition -0.90 0.28 -0.25**
DF Schema -0.13 0.17 -0.07
Condition x DF Schema -0.16 0.31 -0.05 0.00

Note. ASA = Adult Script Assessment, AB = YSQ Abandonment Schema, MA = YSQ Mistrust/Abuse Schema,
ED = Emotional Deprivation Schema, SI = YSQ Social Isolation Schema, and DF = YSQ Defectiveness
Schema. * p < .05. ** p < .01.

Table 15. Full Summary of Repeated Measures Analysis of Variance for Adult Script Assessment
(ASA) Predicting Change in Affective Quality of Representations

Effect MS df F p Partial η2
Time 0.72 1 8.05 0.01 0.05*
Time x Condition 0.09 1 0.99 0.32 0.01
Time x Secure 0.29 1 3.21 0.08 0.02
Time x ASA 0.06 1 0.63 0.43 0.00
Time x Cond x Secure 0.23 1 2.53 0.11 0.02
Time x Secure x ASA 0.06 1 0.62 0.43 0.00
Time x Cond x ASA 0.01 1 0.05 0.82 0.00
Time x Cond x Secure x ASA 0.73 1 8.12 0.01 0.05*
Error 0.09 175
Time 0.34 1 3.61 0.06 0.02
Time x Condition 0.24 1 2.49 0.12 0.01
Time x Fearful 0.00 1 0.01 0.93 0.00
Time x ASA 0.00 1 0.02 0.88 0.00
Time x Cond x Fearful 0.07 1 0.74 0.39 0.00
Time x Fearful x ASA 0.05 1 0.48 0.49 0.00
Time x Cond x ASA 0.06 1 0.62 0.43 0.00
Time x Cond x Fearful x ASA 0.04 1 0.37 0.54 0.00
Error 0.10 171
Time 0.14 1 1.45 0.23 0.01
Time x Condition 0.26 1 2.72 0.10 0.02
Time x Preoccupied 0.04 1 0.40 0.53 0.00
Time x ASA 0.01 1 0.15 0.70 0.00
Time x Cond x Preoccupied 0.05 1 0.54 0.46 0.00
Time x Preoccupied x ASA 0.00 1 0.04 0.84 0.00
Time x Cond x ASA 0.14 1 1.43 0.23 0.01
Time x Cond x Preoccupied x ASA 0.13 1 1.35 0.25 0.01
Error 0.10 171

86
Table 15 (continued).

Time 0.01 1 0.14 0.71 0.00

Time x Condition 0.02 1 0.19 0.66 0.00
Time x Dismissing 0.24 1 2.53 0.11 0.02
Time x ASA 0.00 1 0.03 0.87 0.00
Time x Cond x Dismissing 0.04 1 0.41 0.52 0.00
Time x Fearful x ASA 0.03 1 0.31 0.58 0.00
Time x Cond x ASA 0.17 1 1.76 0.19 0.01
Time x Cond x Dismissing x ASA 0.15 1 1.60 0.21 0.01
Error 0.09 171

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

Table 16. Full Summary of Repeated Measures Analysis of Variance for Adult Script Assessment
(ASA) Predicting Change in Experience and Management of Aggressive Impulses

Effect MS df F p Partial η2
Time 0.53 1 10.35 0.00 0.06*
Time x Condition 0.04 1 0.70 0.40 0.00
Time x Secure 0.01 1 0.16 0.69 0.00
Time x ASA 0.00 1 0.06 0.81 0.00
Time x Cond x Secure 0.33 1 6.46 0.01 0.04*
Time x Secure x ASA 0.01 1 0.19 0.67 0.00
Time x Cond x ASA 0.02 1 0.33 0.57 0.00
Time x Cond x Secure x ASA 0.30 1 5.95 0.02 0.03*
Error 0.05 171
Time 0.25 1 4.61 0.03 0.03*
Time x Condition 0.08 1 1.49 0.23 0.01
Time x Fearful 0.11 1 2.01 0.16 0.01
Time x ASA 0.02 1 0.29 0.59 0.00
Time x Cond x Fearful 0.03 1 0.52 0.47 0.00
Time x Fearful x ASA 0.03 1 0.52 0.47 0.00
Time x Cond x ASA 0.04 1 0.76 0.38 0.00
Time x Cond x Fearful x ASA 0.05 1 0.87 0.35 0.01
Error 0.05 171
Time 0.68 1 13.02 0.00 0.07
Time x Condition 0.21 1 4.01 0.05 0.02
Time x Preoccupied 0.15 1 2.80 0.10 0.02
Time x ASA 0.00 1 0.00 0.97 0.00
Time x Cond x Preoccupied 0.10 1 1.84 0.18 0.01
Time x Preoccupied x ASA 0.02 1 0.36 0.55 0.00
Time x Cond x ASA 0.06 1 1.19 0.28 0.01
Time x Cond x Preoccupied x ASA 0.11 1 2.03 0.16 0.01
Error 0.05 171
Time 0.34 1 6.20 0.01 0.04*
Time x Condition 0.06 1 1.17 0.28 0.01
Time x Dismissing 0.01 1 0.14 0.71 0.00
Time x ASA 0.00 1 0.00 0.95 0.00
Time x Cond x Dismissing 0.00 1 0.03 0.87 0.00

87
Table 16 (continued).

Time x Fearful x ASA 0.00 1 0.00 1.00 0.00

Time x Cond x ASA 0.02 1 0.27 0.61 0.00
Time x Cond x Dismissing x ASA 0.01 1 0.15 0.70 0.00
Error 0.06 171

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

Table 17. Full Summary of Repeated Measures Analysis of Variance for Adult Script Assessment
(ASA) Predicting Change in Self-Esteem

Self-Esteem – preoccupied style contrast

Effect MS df F p Partial η2
Time 0.02 1 1.26 0.26 0.01
Time x Condition 0.04 1 2.78 0.10 0.02
Time x Secure 0.06 1 4.13 0.04 0.02*
Time x ASA 0.01 1 0.35 0.56 0.00
Time x Cond x Secure 0.01 1 0.92 0.34 0.01
Time x Secure x ASA 0.02 1 1.16 0.28 0.01
Time x Cond x ASA 0.00 1 0.01 0.92 0.00
Time x Cond x Secure x ASA 0.02 1 1.17 0.28 0.01
Error 0.02 171
Time 0.00 1 0.08 0.78 0.00
Time x Condition 0.03 1 1.92 0.17 0.01
Time x Fearful 0.01 1 0.41 0.52 0.00
Time x ASA 0.04 1 3.03 0.08 0.02
Time x Cond x Fearful 0.02 1 1.44 0.23 0.01
Time x Fearful x ASA 0.18 1 12.67 0.00 0.07*
Time x Cond x ASA 0.02 1 1.52 0.22 0.01
Time x Cond x Fearful x ASA 0.02 1 1.19 0.28 0.01
Error 0.01 171
Time 0.00 1 0.13 0.72 0.00
Time x Condition 0.08 1 5.15 0.02 0.03*
Time x Preoccupied 0.00 1 0.13 0.72 0.00
Time x ASA 0.00 1 0.29 0.59 0.00
Time x Cond x Preoccupied 0.03 1 1.79 0.18 0.01
Time x Preoccupied x ASA 0.00 1 0.00 0.99 0.00
Time x Cond x ASA 0.01 1 0.45 0.50 0.00
Time x Cond x Preoccupied x ASA 0.00 1 0.21 0.65 0.00
Error 0.02 171
Time 0.00 1 0.08 0.78 0.00
Time x Condition 0.07 1 44.2 0.04 0.03*
Time x Dismissing 0.03 1 2.06 0.15 0.01
Time x ASA 0.05 1 3.19 0.08 0.02
Time x Cond x Dismissing 0.03 1 1.93 0.17 0.01
Time x Fearful x ASA 0.08 1 5.53 0.02 0.03
Time x Cond x ASA 0.00 1 0.07 0.80 0.00
Time x Cond x Dismissing x ASA 0.00 1 0.02 0.89 0.00

88
Table 17 (continued).

Error 0.02 171

Note. ASA = Adult Script Assessment. * p < .05, ** p < .01.

89