ArticlePDF Available

Abstract

Recent developments in pattern analysis research have made this methodology suitable for the study of the processes that are set in motion in psychological interventions. Outcome research, based on the comparison between clinical results from treatment and control groups, has leveraged our empirical knowledge about the efficacy of psychological interventions. However, these methods of research are not precise enough for the analysis of these processes. On the contrary, pattern analysis could be a powerful tool to study moment-to-moment interactions typical of psychological interventions. This is methodology is relevant because clinical psychology is experiencing a paradigm shift from a protocol for syndrome perspective to a principle-based and person-centered intervention. This evidence-based, theory-grounded, and process-oriented paradigm of clinical intervention needs new research methods to thrive (i.e., pattern analysis). The analysis of the therapeutic relationship built into the verbal interaction between the clinician and the client is one of the cornerstones of this new era of research. So, the purpose of this article is three-fold: (1) to discuss the role of the verbal interaction pattern analysis in the clinical context to the development of the principle-based clinical psychology, (2) to analyze the patterns of verbal interaction in a clinical case, and (3) to compare the results using two different methods. To reach these purposes, using the observational methodology, we have coded the verbal interaction of 16 clinical sessions with a person diagnosed with a borderline personality disorder. We have analyzed the data using sequential analysis (GSEQ) and pattern recognition algorithms (i.e., T-Pattern detection). We have been able to detect typical patterns during different phases of psychological intervention (i.e., evaluation, explanation, treatment, and consolidation). Finally, the conceptual, methodological, and empirical implications of this study will be discussed within the realms of pattern analysis research and principle-based clinical psychology.
Frontiers in Psychology 01 frontiersin.org
Verbal interaction pattern
analysis in clinical psychology
JesúsAlonso-Vega           
1,2*, NataliaAndrés-López
2 and
MaríaXesúsFroxán-Parga
2
1 Faculty of Biomedical and Health Sciences, Universidad Europea de Madrid, Madrid, Spain,
2 Department of Biological and Health Psychology, Universidad Autónoma de Madrid, Madrid, Spain
Recent developments in pattern analysis research have made this methodology
suitable for the study of the processes that are set in motion in psychological
interventions. Outcome research, based on the comparison between clinical
results from treatment and control groups, has leveraged our empirical knowledge
about the ecacy of psychological interventions. However, these methods of
research are not precise enough for the analysis of these processes. On the
contrary, pattern analysis could bea powerful tool to study moment-to-moment
interactions typical of psychological interventions. This is methodology is relevant
because clinical psychology is experiencing a paradigm shift from a protocol for
syndrome perspective to a principle-based and person-centered intervention.
This evidence-based, theory-grounded, and process-oriented paradigm of
clinical intervention needs new research methods to thrive (i.e., pattern analysis).
The analysis of the therapeutic relationship built into the verbal interaction
between the clinician and the client is one of the cornerstones of this new era of
research. So, the purpose of this article is three-fold: (1) to discuss the role of the
verbal interaction pattern analysis in the clinical context to the development of the
principle-based clinical psychology, (2) to analyze the patterns of verbal interaction
in a clinical case, and (3) to compare the results using two dierent methods. To
reach these purposes, using the observational methodology, wehave coded the
verbal interaction of 16 clinical sessions with a person diagnosed with a borderline
personality disorder. Wehave analyzed the data using sequential analysis (GSEQ)
and pattern recognition algorithms (i.e., T-Pattern detection). Wehave been able
to detect typical patterns during dierent phases of psychological intervention
(i.e., evaluation, explanation, treatment, and consolidation). Finally, the conceptual,
methodological, and empirical implications of this study will bediscussed within
the realms of pattern analysis research and principle-based clinical psychology.
KEYWORDS
process-based therapy, clinical psychology, pattern analysis, sequential analysis,
verbal interaction, principle-based therapy
Introduction
Pattern is a term that refers to a stable repetition of events that arise from specic
circumstances. e recognition of patterns in the natural world has had an evolutionary
impact on the animal species, it has allowed us to adapt to the environment by taking
advantage of these regularities. Moreover, understanding how these patterns work has led
TYPE Original Research
PUBLISHED 26 July 2022
DOI 10.3389/fpsyg.2022.949733
OPEN ACCESS
EDITED BY
Gudberg K. Jonsson,
University of Iceland,
Iceland
REVIEWED BY
Alain Blanchet,
Ecole de Psychologues Praticiens (EPP),
France
Valentino Zurloni,
University of Milano-Bicocca, Italy
*CORRESPONDENCE
Jesús Alonso-Vega
jesus.alonso@universidadeuropea.es
SPECIALTY SECTION
This article was submitted to
Quantitative Psychology and Measurement,
a section of the journal
Frontiers in Psychology
RECEIVED 21 May 2022
ACCEPTED 27 June 2022
PUBLISHED 26 July 2022
CITATION
Alonso-Vega J, Andrés-López N and
Froxán-Parga MX (2022) Verbal interaction
pattern analysis in clinical psychology.
Front. Psychol. 13:949733.
doi: 10.3389/fpsyg.2022.949733
COPYRIGHT
© 2022 Alonso-Vega, Andrés-López and
Froxán-Parga. This is an open-access
article distributed under the terms of the
Creative Commons Attribution License
(CC BY). The use, distribution or
reproduction in other forums is permitted,
provided the original author(s) and the
copyright owner(s) are credited and that
the original publication in this journal is
cited, in accordance with accepted
academic practice. No use, distribution or
reproduction is permitted which does not
comply with these terms.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 02 frontiersin.org
us to control and predict these natural events. Recognition and
analysis of patterns are scientic endeavors that have beneted
from the eorts of multiple scientic elds (e.g., mathematical
denitions, biological theories, etc.) and are useful to improve
applied sciences (e.g., engineering, medicine, etc.). Furthermore,
the development of new denitions of patterns or the application
of existing methods of analysis to new areas could solve
contemporary problems in several scientic disciplines, webelieve
that this is the case in clinical psychology. is paper aims to
improve clinical psychology research using pattern analysis
methods (e.g., T-pattern Analysis), to do that rst wedene the
state of the art in clinical psychology research highlighting the
current challenges, then weanalyze the conceptual and practical
opportunities that pattern analysis methods open to this eld of
research, and nally weconduct a proof-of-concept study.
Clinical psychology has reached an important stage of
development in regards to the scientic agreement about the
ecacy of psychological interventions for specic psychological
problems (David etal., 2018; Van Agteren etal., 2021). anks to
the evidence-based psychotherapy movement, this milestone has
ended a historical debate within our eld (Forte etal., 2014).
While, it has set the opportunity to face new challenges and
research questions, as well as analyze the aws of the current
clinical research system. For example, Paul’s classic question,
“What treatment, by whom, is most eective for this individual
with that specic problem, and under which set of circumstances,
and how does it come about?” (Paul, 1969, p.44), is not yet
solved. To answer this question weneed to implement conceptual,
experimental, and applied changes to our clinical research and
our validation systems (Tolin etal., 2015). Among these changes,
wewould like to emphasize the importance of clinical process
research rooted in a functional perspective of behavior,
experimentally validated processes, and the connection between
intervention outcomes and the processes that could explain these
behavioral changes (Hayes and Hofmann, 2018; Barnes-Holmes
etal., 2020). Especially, two main trends are rising in clinical
psychology, the call for personalized psychotherapy and the study
of complex interactions between psychological problems,
psychological interventions, and treatment outcomes (Hofmann
etal., 2020; Wright and Woods, 2020; Mansueto etal., 2022). Due
to the nature of these two trends, clinical research needs new
methods and approaches to conduct idiographic studies in
clinical psychology. ere is a long tradition of idiographic
research in clinical psychology (Kazdin, 1978; Hayes, 1981; Iwata
etal., 1994), but the emphasis on this type of research is renewed
(e.g., Molenaar, 2004; Beltz etal., 2016; Piccirillo etal., 2019;
Kazdin, 2021). New professional standards for single-case
methods (Tate etal., 2016; Horner and Ferron, 2022) and the
development of new methods for the interpretation of these
designs (Hedges et al., 2013; Kratochwill and Levin, 2014;
Pustejovsky, 2018; Manolov etal., 2021) could help us to analyze
the learning processes, that occurs at the individual level and are
settled in motion by the psychologists’ procedures, that account
for the clients change.
Specically, in the psychological interventions with adults,
these processes occurred within the verbal interaction between
the psychologist and the client. Other important variables explain
the therapeutic change of our clients (e.g., contextual, and cultural
variables outside of the clinical session, setting variables,
motivational variables, etc.), but the verbal interaction with the
client is the main channel that psychologists have to implement
their procedures (Follette and Bonow, 2009; Tsai et al., 2014;
Virues-Ortega and Froxán-Parga, 2015).
e interest in the study of verbal interaction in clinical
sessions from a functional perspective is present since the early
60 s (see Moore, 1991). is conceptual development helped to
foster empirically-supported psychological treatments like the
Functional Analytic Psychotherapy (Kohlenberg and Tsai, 1991)
or the Acceptance and Commitment erapy (ACT; Hayes etal.,
1999; Hayes, 2004)
1
. Also, helped to develop coding systems for
the observational study of the verbal interaction like the
Functional Analytic Psychotherapy Rating Scale (FAPRS;
Callaghan etal., 2008); the Multidimensional System for Coding
Behaviors in erapist-Client interaction (SiMCCIT; Zamignani,
2008) or the Verbal Interaction Categorization System in erapy
(Froxán-Parga etal., 2011; Alonso-Vega etal., 2022). Using these
coding systems allowed to describe the verbal interaction
between psychologist and client in clinical sessions (e.g., Froxán-
Parga etal., 2016), to study the eects of it outside of the session
(e.g., Lizarazo et al., 2015), to study the molecular learning
processes that occur in this interaction (e.g., Busch etal., 2009),
to analyze the interaction during specic techniques like
cognitive restructuring (e.g., Calero-Elvira etal., 2013; Froxán-
Parga etal., 2018), and to conduct experiments to study the
verbal shaping during clinical sessions (Pardo-Cebrian et al.,
2021). ese works helped to analyze the basic principles of
change in psychological interventions, however, they report
methodological problems to study moment-to-moment
interactions. For example, Busch etal. (2009) and Xavier etal.
(2012) used transitional probabilities to study these interactions.
Transitional probabilities inform us about the probability of an
event (X) given other (Y), but this is a limited method to study
the verbal interaction in psychological interventions because
these probabilities reect behavior frequencies in a particular
session and are not comparable over sessions (Bakeman and
Quera, 2011). Other, studies (see Calero-Elvira etal., 2013) opted
to use sequential analysis to analyze these moment-to-moment
interactions. Sequential analysis techniques help us to study
patterns and temporal associations among behaviors within
observational sessions (Bakeman and Quera, 2012), these
techniques are based on the calculation of contingency indices
1 To analyze the relationship between the early functional
conceptualizations of the verbal behavior to the development of empirically
supported psychological treatments is beyond the scope of this paper,
but it can befound elsewhere (see Hayes, 2004; Froxán-Parga etal., 2018;
Hayes and Hofmann, 2018; Barnes-Holmes etal., 2020).
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 03 frontiersin.org
for 2 × 2 tables so they are limited to two-term contingencies.
Also, these sequential analyses, normally are used to conrm an
expected sequential relationship between two events and are not
used to explore patterns that are hidden from the observer’s eyes
(Magnusson, 2005). us, more sophisticated pattern detection
and analysis techniques could help us to detect hidden patterns
of verbal interaction in clinical sessions and thus to have more
precise data that enable us to better study the learning principles
that could explain the clients behavioral changes that are set in
motion in the verbal interaction between the psychologist and the
client. So, the application of the T-Pattern, using THEME
soware that allows us the automatic detection of temporal and
sequential structures in observational data, could beuseful to
overcome the methodological limitations of previous studies of
the verbal interaction analysis in clinical settings. Consequently,
the purpose of this study is to analyze the patterns of verbal
interaction in a clinical case using the THEME soware and to
compare the results with previously used analyses, as in Lapresa
etal. (2013).
Materials and methods
Participants
For this observational study wehad the participation of a
31-year-old client diagnosed with borderline personality disorder
(BPD) for the last 8 years; and a 35-year-old clinical psychologist
(a master’s degree in General Health Psychology and a master’s
degree in Behavior erapy). Both participants came from a
public-funded Vocational Rehabilitation and Employment center
(VR&E) in the Community of Madrid. e client has been
referred to this center due to mood problems (i.e., emotional
lability) and substance use problems that directly interfere with
the client’s chances of accessing employment/training
opportunities; and once obtained, problems in keeping his job or
completing the required training. Also, this study involved the
participation of two trained observers. Both observers are
predoctoral students that have been trained in the same research
group (i.e., the ACOVEO research group). Observer 1 and
Observer 2 have, respectively, 4 and 2 years of experience working
with the observational coding system used in this research and
they helped in the development of it. Before recording the clinical
session, the client and clinician have been informed about the use
of the data and the purpose of the research. All participants have
signed the study informed consent.
Instruments
We used the Functional Coding System for Verbal
Interaction in Clinical Contexts (Alonso-Vega etal., 2022) to
code the verbal interaction between the client and the
psychologists. is coding system, which focuses on the putative
functions of the verbal behavior, has ve coding categories for
the clinician verbalizations: Clinical Discriminative Stimulus
(CD) and Instructional Discriminative Stimulus (ID),
Conditioned Motivating Operation (CMO), and Positive
Reinforcer (R+) and Aversive Stimulus (AS). e observational
system assumes that the verbal behavior of the client has a
response function it is not established any specic categories in
the observational system. To code the client’s verbal behavior in
this case we used eight categories based on the response
topography Giving Information (GI), Asking for Information
(AI), Following Instructions (FI), Not Following Instructions
(NFI), Well-Being (WB), Discomfort (D), Target Behavior (TB)
and Problem Behavior (PB). See Table1 for a brief description of
the coding categories, also more details of the coding system are
available in the additional materials.
Materials
Recordings of 16 clinical sessions were obtained through
the camera installed on the VR&E psychologist’s computer. e
recordings were sent to the ACOVEO research group and were
treated following the protocol used in the research group in
which the recordings are anonymized following the ethical and
legal guidelines of the Organic Law 3/2018 on Personal Data
Protection and guarantee of digital rights. ese recordings
were stored on external hard drives kept under lock and key in
the group’s laboratory at the Autonomous University
of Madrid.
e recording of the clinical sessions, the observation project,
and the analysis of inter-observer reliability were carried out with
e Observer XT 12 observation soware. e data analysis was
done in R (RStudio Team, 2020) and Microso Excel. GSEQ
(Bakeman and Quera, 2016) was the soware used to conduct the
sequential analysis of the data. Weselected this soware because
it is generally employed in observational research (e.g., Santoyo
etal., 2017; Brown etal., 2018); it was specically used in previous
research in the study of verbal interaction in clinical cases (Calero-
Elvira etal., 2013; Pardo-Cebrian etal., 2021); and it was specially
developed to calculate sequential patterns in observational data
(Bakeman and Quera, 1995, 2016). Finally, weused emeEdu
soware (Pattern Vision, 2021) for automatic pattern detection.
emeEdu was selected because it has been successfully applied in
dierent research areas (i.e., neuronal interactions, behavioral
interactions, etc.; Magnusson, 2020), but it was not employed in
the study of verbal interactions in clinical settings. Both programs
allowed us to conduct the data analysis described in the section
below (see “Data Analysis”).
Procedure
In this observational study, weused an intra-subject design
with three dierent phases: Evaluation (EVA.), Treatment
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 04 frontiersin.org
Phase 1 (T1), and Treatment Phase 2 (T2). ese phases were
not experimentally manipulated and have been divided
considering the protocols and procedures of the VR&E center
(i.e., EVA, rst three sessions; TP1, 4–9 sessions; and TP2,
10–16 sessions). e division between the two treatment
phases is arbitrary and responds to a need to divide the
treatment into, at least, two phases to evaluate dierences
between various time points.
Aer the study, both observers received specic training in
the observational instrument. e training process was completed
when a stable reliability index (k > 0.70) was achieved while coding
similar clinical sessions. ese sessions were from the research
teams clinical sessions archive.
Aer training was completed, observer 1 individually
recorded all treatment sessions. Observer 2, also individually,
recorded 4 random sessions out of the 16 treatment sessions,
representing 25% of the sample, which is above the usual 10% for
studies of this type. e inter-observer reliability calculation was
carried out aer the end of the recording of both observers. One
month aer the end of the recording phase, observer 1 recorded
again two randomly selected treatment sessions to allow the
calculation of intra-observer reliability.
Data analysis
e kappa coecient (k) was used to calculate the inter-
observer and intra-observer reliability of the records. Once the
records were obtained, the rate per minute of each variable
recorded in each session was calculated. Descriptive data (e.g.,
count, rate per minute, etc.) were obtained to allow a visual
inspection of the variables and the patterns through time. To
analyze the interaction between variables, weused two types of
statistical analysis, which are part of the family of statistical tools
for the sequential analysis of temporally distributed data. First,
wecalculated the Yule’s Q, a contingency index of 2×2 tables, using
the GSEQ soware (Bakeman and Quera, 1995, 2016). is index
allows us to calculate the Lag +1 correlation between a given
behavior and the one that follows it. e Lag-1 correlation tells us
which behavior precedes the behavior weare analyzing. is index
allows descriptive and analytical analysis of the association
(Bakeman and Quera, 2016) and its scores can also beinterpreted
in the same way as Cohen’s r. To study the association between
specic pairs of behaviors wecalculated the adjusted residuals (z),
which are a normalized index of the extent to which the values of
the frequencies observed in each cell of the matrix deviate from
their expected values: a value greater than 1.96 indicates that this
behavior occurs signicantly more than expected and, conversely,
a value less than 1.96 implies that it occurs signicantly less than
expected by chance (p < 0.05; see Bakeman and Quera, 2011 for an
advance mathematical description).
Moreover, we used a pattern recognition model T-Pattern
Model (Magnús et al., 2016; Casarrubea et al., 2018), using
emeEdu soware (Pattern Vision, 2021). is model allows us
to recognize patterns (T-Patterns) from observational data
(T-Data) using the T-Pattern detection algorithm. Pattern
detection works in a bottom-up fashion, from the data to the
pattern detection. e T-Pattern is a hierarchical, multi-ordinal,
and self-similar pattern type that comprises m ordered
components (i.e., behavioral events), X1.m, recurring in a single
discrete dimension, where each component is a T-data category
(or pattern primitive, called event-type) or a T-pattern
(Magnusson, 2020).
In this case, patterns of interaction between client and therapist
have been detected in all three phases of observation. Weused the
pattern recognition default settings with some specications.
Wehave required that the patterns must berepeated at least three
times in each session and all sessions of each phase, with this
wewant to make sure that the pattern is a characteristic of this
phase not a characteristic of one of the sessions. Also, wehave
excluded the patterns made by the repetition of the same variable
and the patterns made by an interaction of variables of the same
subject, because we want to study the interaction between
variables. Finally, wehave required a maximum of 4 levels for the
TABLE1 Brief description of the coding categories.
Coding categories Abbreviation Putative function Example
Clinical Discriminative CD Antecedent stimulus that increases the probability of a response class.
(e.g., to give clinical information)
“How do youfeel about that?
Instructional Discriminative ID Antecedent stimulus that increases the probability of a response class
(e.g., to follow instructions)
“I want youto apply the breathing
technique every night.
Conditioned Motivating
Operation
CMO Antecedent stimulus that changes the reinforcing value of a consequent
stimulus and changes the frequency of responses related with this
consequent stimulus
“If yourun daily, your situation will
i m p r o v e .”
Positive Reinforcer R+ Consequent stimulus that increases the probability of an operant
response with a positive contingency with it
“Very good! Youhave done great.
Aversive Stimulus AS Consequent stimulus that decreases the probability of an operant
response with a positive contingency
“I do not agree with what youhave done.
is table only displays a brief description of the clinician coding categories. Client coding categories are topographically based and they depend on each clinical case. Please see the
additional materials for further details.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 05 frontiersin.org
analysis of the interaction, because more than four levels could
be interpreted as chains made of contingencies of two or
three members.
To assess the eect of the treatment on the client’s behaviors
wecalculated the eect size, in this study weused the Non-overlap
of All Pairs (NAP) which is an index focused on identifying the
dierences between two phases of a design (A and B; Parker and
Vannest, 2009; Carter, 2013).
Results
Reliability
Inter-rater reliability is 0.71–0.86, and intra-rater reliability is
0.82–0.89. Both k values can be interpreted as very good and
excellent reliability indexes (McHugh, 2012).
Using SDIS-GSEQ
Figures 1, 2 show the results of antecedent (Lag+1) and
consequent (Lag-1) sequential analysis using GSEQ. All
displayed correlations are signicant (p < 0.01), positive
correlations (Q > 0) are painted in green, and negative
correlations (Q < 0) are painted in red. Data indicate that there is
a positive correlation between Clinical Discriminative Stimulus
with the client’s Giving Information behavior in all phases of the
case (see Figure1). Following Instructions correlates positively
with the occurrence of Instructional Discriminative Stimulus, but
this only occurs in the treatment phases (i.e., it is not found in
the assessment phase); Following Instructions correlates
positively with Target Behaviors during Evaluation and T1.
Finally, Figure 1 also shows a signicant positive correlation
between Conditioned Motivating Operation and Target Behaviors
in all phases.
e consequent sequential analysis indicates a positive
correlation between Positive Reinforcers and Target Behaviors,
Well-being Verbalizations, Following Instructions, Aversive Stimuli,
and Problem Behavior. Also Conditioned Motivating Operations
were weakly correlated with Target Behaviors and Asking
for Information.
Using THEME
Tables 2, 3 show the frequency of two-, three- and four-
term patterns in each part of the clinical case. For example,
Table 2 displays a strong relationship between Clinical
Discriminative and Giving Information in all phases of the
clinical case, this pattern repeats 888 times through dierent
clinical sessions. Also, there are some patterns especially
repeated in the early stages of the treatment (e.g., Clinical
Discriminative and Discomfort, Instructional Discriminative
and Target Behavior), and in treatment phases (e.g., Clinical
Discriminative and Wellbeing, Clinical Discriminative, and
Target Behaviors). Moreover, we have detected patterns
increasing through the phases (e.g., Conditioned Motivating
Operation and Target Behaviors, Target behaviors and
Positive Reinforcers).
Table3 shows the three- and four-term patterns in each phase.
e most repeated three-term patterns involve Target Behaviors
and Positive Reinforcers (e.g., DC TB R+; CMO TB R+). Our
analysis reveals similar data with four-term patterns. e most
repeated patterns involve Target Behaviors and Positive
Reinforcement (e.g., GI DC TB R+; TB CMO TB R+).
FIGURE1
Antecedent sequential analysis using GSEQ. This figure displays
sequential relationships between clinician’s behaviors and the
client’s behaviors that have been followed by. Here the positive
relationships are shown in green and indicate that these are
significatively correlated behaviors (Q > 0; p < 0.01.). For example,
CD was followed by GI in all treatment phases. Negative ones, in
red, indicate that those behaviors were not observed together
during the verbal interaction (Q < 0; p < 0.01). CD, Clinical
Discriminative Stimulus; ID, Instructional Discriminative Stimulus;
CMO, Conditioned Motivating Operation; GI, Giving Information;
TB, Target Behavior; PB, Problem Behavior; FI, Following
Instructions; Ti, Treatment phase 1; T2, Treatment phase 2.
FIGURE2
Consequent sequential analysis using GSEQ. This figure displays
sequential relationships between client’s behaviors and the
clinician’s behaviors that have been followed by. Here the
positive relationships are shown in green and indicate that these
are significatively correlated behaviors (Q > 0; p < 0.01). For
example, TB was followed by R+ in all treatment phases.
Negative ones, in red, indicate that those behaviors were not
observed together during the verbal interaction (Q < 0; p < 0.01).
R+, Positive Reinforcer; AS, Aversive stimulus; CMO, Conditioned
Motivating Operation; GI, Giving Information; TB, Target
Behavior; WB, Well-Being; Problem Behavior; FI, Following
Instructions; AI, Asking for Information; Ti, Treatment phase 1; T2,
Treatment phase 2.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 06 frontiersin.org
TABLE3 Three- and four-term pattern analysis using THEME.
Pattern type Number of patterns
ree-term
pattern
Evaluation T1 T2
CD-GI-R+ 15
CD-TB-R+ - 95
ID-TB-CMO 9 15 -
CMO -TB-R+ 31 35 66
D-CD-GI 9
WB-CD-GI 12
PB-AS-TB 10
Four-term pattern
GI-DC-TB-R+ 55
TB-CMO-TB R+ 21
DC-TB-R + -PB 9
CD, Clinical Discriminative Stimulus; ID, Instructional Discriminative Stimulus; CMO,
Conditioned Motivating Operation; R+, Positive Reinforcer; AS, Aversive stimulus; GI,
Giving Information; TB, Target Behavior; PB, Problem Behavior; T1, Treatment phase 1;
T2, Treatment phase 2; p < 0.01.
Comparison between both software
Table4 shows a comparison between the patterns with two
variables detected with GSEQ and THEME. Specically, 59% (i.e.,
17/29) of the identied patterns were detected by both methods
of detection. While 31% (i.e., 9/29) were detected only by THEME,
10% (i.e., 3/29) were detected by GSEQ.
Visual distribution of the patterns
Pattern detection has indicated which variables are
moment-to-moment correlated in verbal interaction. ese
results have allowed us to analyze how these variables change
along with the psychological treatment. Figures3, 4 show the
distribution, through a clinical session, of the variables that are
present in the most repeated patterns. ese gures have
been useful for visual analysis of the covariation of
correlated variables.
For example, Figure 3 displays the evolution of
verbalizations with Clinical Discriminative functions among
different treatment phases and how are they followed by
Giving information verbalizations of the client. As we will
discuss these visual inspections could help practitioners to
evaluate the clinical relationship or the efficacy of the
clinical intervention.
TABLE2 Two-term pattern analysis using THEME.
Pattern Number of patterns
Clinician Client Evaluation T1 T2
CD GI 221 424 243
CD WB 24 13
CD TB 81
CD AI 19
CD PB 27
CD D 10
ID TB 15 19
CMO TB 54 87 104
CMO AI 14
Client Clinician Evaluation T1 T2
TB CMO 41 25
TB R+ 65 63 97
WB DC 16
WB R+ 11 21
GI R+ 19 56
PB AS 14 14
FI R+ 13
AI CMO 10
CD, Clinical Discriminative Stimulus; ID, Instructional Discriminative Stimulus; CMO,
Conditioned Motivating Operation; R+, Positive Reinforcer; AS, Aversive stimulus; GI,
Giving Information; WB, Well-Being; AI, Asking for Information; TB, Target Behavior;
FI, Following Instructions; PB, Problem Behavior; T1, Treatment phase 1; T2, Treatment
phase 2; p < 0.01.
TABLE4 Two-term patterns detected in each software.
Antecedent patterns Evaluation T1 T2
CD-GI GSEQ and
THEME
GSEQ and
THEME
GSEQ and
THEME
CD-WB THEME THEME
CD-TB THEME
CD-AI THEME
CD-PB THEME
ID-FI GSEQ GSEQ
ID-TB GSEQ and
THEME
GSEQ and
THEME
CMO-TB GSEQ and
THEME
GSEQ and
THEME
GSEQ and
THEME
CMO-AI THEME
Consequent patterns
TB-R+ GSEQ and
THEME
GSEQ and
THEME
GSEQ and
THEME
TB-CMO THEME GSEQ and
THEME
WB-R+ GSEQ and
THEME
GSEQ and
THEME
PB-AS GSEQ GSEQ and
THEME
GSEQ and
THEME
AI-CMO GSEQ and
THEME
GI-R+ THEME
FI-R+ THEME
Antedecent patterns are formed by clinician’s behaviors followed by the client’s
behaviors; Consequent patterns are formed by clinician’s behaviors followed by the
client’s behaviors; CD, Clinical Discriminative Stimulus; ID, Instructional Discriminative
Stimulus; CMO, Conditioned Motivating Operation; R+, Positive Reinforcer; AS,
Aversive stimulus; GI, Giving Information; WB, Well-Being; AI, Asking for Information;
TB, Target Behavior; FI, Following Instructions; PB, Problem Behavior; T1, Treatment
phase 1; T2, Treatment phase 2; p < 0.01.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 07 frontiersin.org
Eect sizes
Both sequential analysis and pattern analysis indicate
positive correlations between the client’s Target Behaviors and
the therapist’s Positive Reinforcers. To assess the effect of these
Positive Reinforcers wehave calculated the effect size in the
client’s Target Behaviors (e.g., Positive Reinforcers should
increase the likelihood of emission of the behaviors they
follow). In this case, the effect size index indicates that the
treatment had a moderate effect on increasing the Target
Behaviors (NAP = 0.67; Standard Error: 0.21; 95% CI:
[0.32–0.89]).
Discussion
e aim of the current article was three-fold. First, wehave
highlighted the potential relevance of pattern analysis methods in
the study of verbal interaction in psychological interventions and
wehave presented the reasons why these types of analysis could
help in the development of the next generation of process analysis
in clinical research. Second, wehave conducted an observational
proof-of-concept study to analyze the verbal interaction in a
single-case design using two methods of pattern analysis (e.g.,
GSEQ and THEME). Using sequential analysis and more complex
pattern detection algorithms wecould identify more than 25
interaction patterns between a clinical psychologist and his client
in dierent phases of the psychological treatment. ese
interaction patterns were visually displayed through dierent
sessions, and the eect size of the treatment was measured for the
Target Behaviors of the client. ird, wehave compared the results
yielded by these two methods of pattern analysis. In this section
of the article, we discuss methodological implications in the
search for hidden patterns of interaction in clinical settings, the
clinical implications of our results, the limitations of our work,
and the main conclusions.
Methodological implications
We have compared the performance of GSEQ and THEME
soware in the detection of patterns in verbal interaction. ey
provided similar results in the detection of sequential patterns of
two variables; approximately 60% of these patterns were detected
by both (see Table4). Although 60% could seem a low percentage
of agreement, they detected in the most repeated patterns and
there was not any dierence in the main repeated patterns.
THEME detected 9 patterns, in dierent moments of the
treatment, that GSEQ did not. It seems that THEME could
be more sensible to patterns with lower frequencies than the
GSEQ. us, if the purpose is to explore hidden patterns of
interaction, THEME is more useful in this regard. But if the aim
is to detect the most signicant patterns of interaction between
two variables, both perform equally. At this point, wewould like
to bring attention to the negative correlations exposed by
GSEQ. Both methods could beuseful to assess treatment integrity,
but the negative correlations give us extra information. It gives us
not just about the apparition of psychologist behavior when a
specic client’s behavior occurs, but also the absence of a certain
psychologist behavior aer the client’s behavior. For example,
thanks to this data wecould detect that his psychologist has not
presented any positive reinforcer aer a problem behavior of the
client (see Figure2).
One of the main dierences between both soware is the
detection of patterns constituted by more than two variables.
THEME showed to bepowerful enough to automatically detect
these patterns. While with GSEQ is possible to calculate dierent
lag distances (e.g., Lag +2, Lag +3, etc.), the interpretation could
FIGURE3
Clinical Discriminative and Giving Information distribution though
sessions. This figure displays the rate per minute covariation
between two sequentially correlated behaviors through dierent
clinical sessions. Specifically, here wecan see how the clinician’s
behavior Clinical Discriminative Stimulus rate matches with the
client’s behavior Giving Information during the treatment. CD,
Clinical Discriminative Stimulus; GI, Giving Information; T1,
Treatment phase 1; T2, Treatment phase 2.
FIGURE4
Target Behavior and Positive Reinforcer distribution though
sessions. This figure displays the rate per minute covariation
between two sequentially correlated behaviors through dierent
clinical sessions. Specifically, here wecan see how the clinician’s
behavior Positive Reinforcer rate matches with the client’s
behavior Target Behavior during the treatment. R+, Positive
Reinforcer; TB, Target Behavior.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 08 frontiersin.org
lead to erroneous conclusions, because it did not include the
variables that are inside of this correlation, and the soware simply
correlate two variables that are in a specic distance. at is not the
case with THEME, it automatically has detected patterns formed by
three and four correlated variables. is feature is essential to the
study of verbal interaction because it has been useful to detect the
repetition of structured patterns that imply a use by the psychologist
of three- and four-term contingencies. In this study, we have
detected 7 three-term patterns and 3 four-term patterns. is
detection imply that wehave increased the precision of this analysis,
if wecompare it with previous research on this topic (e.g., Calero-
Elvira etal., 2013). us, the combination of both methods seems
to besuitable to detect negative correlations and to detect complex
patterns (i.e., three- and four-term contingencies).
Clinical implications
As we have discussed, the use of sequential analyses and
pattern recognition algorithms to analyze the verbal interaction
between psychologist and client could help us to study how the
learning processes are set in motion in psychological interventions.
e observational data analysis of this paper has permitted us to
have a closer look at this interaction and to describe how this
interaction has occurred during the treatment. Among the results,
wewould like to discuss the clinical relevance of several patterns.
Specically, wewere able to detect that Clinical Discriminative
Stimulus correlates signicantly with the client’s Giving
Information behavior in all phases of the case, showing antecedent
discriminative control by the psychologist of the client’s Giving
Information behavior. Figure 3 shows how the rate of these
behaviors changes similarly during treatment. ese data could
inform us that the psychologist has a good therapeutical
relationship with the client, due to the positive correlation between
these two variables. e sequential antecedent analysis indicates
that Following Instructions correlates positively with the
occurrence of Instructional Discriminative Stimulus, but this only
occurs in the treatment phases and it is not found in the
assessment phase. ese dierences between the evaluation phases
could be explained by the positive correlation found in the
evaluation phase between Target Behaviors and Instructional
Discriminative Stimulus. is contingency also seems to occur to
a lesser extent in the rst treatment phase and coincides with the
lower correlation between Following Instructions and Instructional
Discriminative Stimulus in the treatment phases. Instructional
Discriminative Stimulus may have correlated in the assessment
with topographies of the Target Behaviors, but this study is not
sensitive to such topographies. Also, wehave detected a correlation
between Target Behaviors and the Conditioned Motivating
Operations. is correlation could beclinically explained by the
conditioning function of the conditioned motivation operations.
ese verbalizations have the purpose of changing the client’s
motivational value of some stimulus or activities. If the clinician
states a verbalization with this function and the client agrees, it is
probable that this agreement could becoded as a target behavior.
Also, if the clients state a Target Behavior, the clinician could
explain more details about why the client is right or relate it to
their therapeutic goals, so these verbalizations could have a
motivating function. Moreover, this relationship between Target
Behaviors and Conditioned Motivating Operations also appears to
have a key role in patterns with more complex structures (i.e.,
three- and four-term patterns; see Table3). Due to this correlation
having the potential impact of changing the client’s value of events,
it could have a signicant role in the clients’ behavioral change
outside the clinical context and it should bestudied in detail in
future studies.
Also, Positive Reinforcers correlate positively with Target
Behaviors, in the three phases of the case; with verbalizations of
Well-Being, in the assessment and the rst part of the treatment;
and with Following Instructions, in the rst part of the treatment.
is could mean that these behaviors are under a schedule of
positive reinforcement applied by the psychologist. At the same
time, positive reinforcers do not correlate with problem behaviors
in the assessment and correlate negatively in the treatment
statements. is could indicate that the therapist identies
problem behaviors once treatment has already begun and hedoes
not apply positive reinforcement schedules. In contrast, Aversive
stimuli correlate positively with Problem Behaviors. us, that
could mean that these behaviors are under a reduction procedure
applied by the psychologist.
As with Conditioned Motivating Operations, Positive Reinforcer
correlates with Target Behaviors even in three- and four-term
patterns. CMO-TB-R+; CD-TB-R+; GI-DC-TB-R+; and TB-CMO-
TB-R+. Figure4 shows how Positive Reinforcers and Target Behaviors
covary through the treatment. It seems that the psychologist was
using reinforcing contingencies that include Target Behaviors. e
theoretical eects of these contingencies should imply an increment
of Target Behavior in session. Calculating the eect size in the
increment of the Target Behaviors, we tried to indirectly assess
whether this procedure aects these behaviors. Results showed us
that the eect size was moderate. Other potential variables could
inuence this class of behaviors, further experimental analysis
should conrm the relationship that wehave detected in this study.
Limitations
We have found problems in the use of the GSEQ and THEME
soware with raw data obtained using e Observer XT 12 to the
GSEQ and THEME soware. Both soware could develop
techniques to import the results from observational soware with
ease. Moreover, as wediscussed, wehave found that the THEMEs
complex pattern detection performance is superior to the GSEQ
performance in the same task. Although THEME seems to
besensible to patterns with low frequency, this is more a challenge
in the interpretation of the results to the researcher than a
limitation of the soware. Also, we have not analyzed all the
patterns that THEME has detected, weimpose some pre-analysis
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 09 frontiersin.org
requirements. Without these requirements, patterns reported by
THEME could increase.
Also, the observational coding of events was not automated.
e functional denitions of the variables have increased the
complexity of the observational coding. In this sense, the
reproducibility of the results is compromised, because this
methodology of research is time-consuming, and it requires
observers with high standards of training. Although the
complexity is an issue, this analysis could bebeneted from the
inclusion of qualitative data on the verbalizations.
Finally, the results of this study are merely tentative and
further experimental analyses need to beconducted to fully study
the patterns of interaction that are occurring in the psychological
interventions. Also, this experimental control could help to better
analyze the eect size of the treatment. e inuence of external
variables could have aected the behavior of the client. Moreover,
werecognize that this study could have all the potential limitations
of the single-case research. For example, the results derived and
analyzed in this study are not representative of the clinical
interaction in all clinical cases, and results could bebiased by
several factors (e.g., culture, psychologists training, client’s
psychological problems, etc.). But, despite all these limitations,
webelieve in the exploratory value of this paper, it could beuseful
for the development of new perspectives and methodologies for
the study of processes in clinical psychology.
Conclusion
e study of the processes underlying therapeutic change is
essential to optimize psychological treatments. e identication
of patterns of verbal interaction during therapy is a valuable step in
understanding how the processes that make psychological
treatments work are set in motion. GSEQ and THEME soware
have proven to beable to detect those patterns in verbal interaction.
THEME has proven to bemore powerful in detecting complex
interaction patterns and more sensitive in detecting low-frequency
patterns, yet both have detected predominant patterns in verbal
interaction that may underlie clinical change. is implies that
pattern recognition methods could be seen as a promising
alternative to studying behavioral change processes in psychological
treatments. ese methods combined with single-case designs and
the development of new recently developed eect sizes for this type
of studies (e.g., Pustejovsky, 2018), could have a unique impact on
the development of clinical research in the forthcoming years.
Data availability statement
e datasets presented in this study can befound in online
repositories. e names of the repository/repositories and
accession number(s) can befound at: https://osf.io/ezhmd/?view_
only=dbe50c17e90a419e93c87116703ceb27.
Ethics statement
e studies involving human participants were reviewed and
approved by Autonomous University of Madrid. e patients/
participants provided their written informed consent to participate
in this study.
Author contributions
All authors listed have made a substantial, direct, and
intellectual contribution to the work and approved it
for publication.
Funding
is research was supported by grant PSI2016-76551-R from
the Ministerio de Economía, Industria y Competitividad of the
Spanish Government.
Conflict of interest
e authors declare that the research was conducted in the
absence of any commercial or nancial relationships that could
beconstrued as a potential conict of interest.
Publisher’s note
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their aliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
References
Alonso-Vega, J., Pereira, G. L., and Froxán-Parga, M. X. (2022). Functional
Coding System for Verbal Interaction in Clinical Contexts. psyarxiv. doi: 10.31234/
osf.io/ytpq5, [Epub Ahead of Print]
Bakeman, R., and Quera, V. (1995). Analyzing Interaction: Sequential Analysis with
SDIS & GSEQ. New York, NY, US: Cambridge University Press.
Bakeman, R., and Quera, V. (2011). Sequential Analysis and Observational
Methods for the Behavioral Sciences. New York, NY: Cambridge University Press.
Bakeman, R., and Quera, V. (2012). “Behavioral observation,” in APA
Handbook of Research Methods in Psychology. Foundations, planning, measures,
and Psychometrics. Vol. 1. eds. H. Cooper, P. M. Camic, D. L. Long, A. T. Panter
and D. Rindskopf (Washington, DC: American Psychological Association),
207–225.
Bakeman, R., and Quera, V. (2016). Generalized Sequential Querier (GSEQ): a
program for Analyzing Sequential data. United States: Gerorgia State University.
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 10 frontiersin.org
Barnes-Holmes, Y., Barnes-Holmes, D., McEnteggart, C., Dougher, M. J., and
Luciano, C. (2020). Introduction to the special issue on clinical psychology and
behavioral science: processes, principles, and analytic strategies. Psychol. Rec. 70,
541–542. doi: 10.1007/s40732-020-00444-9
Beltz, A. M., Wright, A. G., Sprague, B. N., and Molenaar, P. C. (2016). Bridging
the nomothetic and idiographic approaches to the analysis of clinical data.
Assessment 23, 447–458. doi: 10.1177/1073191116648209
Brown, M., Masterson, C., Latchford, G., and Tober, G. (2018). erapist–client
interactions in motivational interviewing: e eect of therapists’ utterances on
client change talk. Alcohol Alcohol. 53, 408–411. doi: 10.1093/alcalc/agy027
Busch, A. M., Kanter, J. W., Callaghan, G. M., Baruch, D. E., Weeks, C. E., and
Berlin, K. S. (2009). A micro-process analysis of functional analytic psychotherapy’s
mechanism of change. Behav. er. 40, 280–290. doi: 10.1016/j.beth.2008.07.003
Calero-Elvira, A., Froxán-Parga, M. X., Ruiz-Sancho, E. M., and
Alpañés-Freitag, M. (2013). Descriptive study of the Socratic method: evidence for
verbal shaping. Behav. er. 44, 625–638. doi: 10.1016/j.beth.2013.08.001
Callaghan, G. M., Follette, W. C., Ruckstuhl, L. E. Jr., and Linnerooth, P. J. N.
(2008). e functional analytic psychotherapy rating scale (FAPRS): A behavioral
psychotherapy coding system. Behav. Analyst Today 9, 98–116. doi: 10.1037/
h0100648
Carter, M. (2013). Reconsidering overlap-based measures for quantitative
synthesis of single-subject data: what they tell us and what they Don’t. Be hav. M odif .
37, 378–390. doi: 10.1177/0145445513476609
Casarrubea, M., Magnusson, M. S., Anguera, M. T., Jonsson, G. K., Castañer, M.,
Santangelo, A., et al. (2018). T-pattern detection and analysis for the discovery of
hidden features of behaviour. J. Neurosci. Methods 310, 24–32. doi: 10.1016/j.
jneumeth.2018.06.013
David, D., Lynn, S. J., and Montgomery, G. eds. (2018). Evidence-based
Psychotherapy: the state of the Science and Practice. 1st Edn. Hoboken, NJ, USA:
Wile y.
Follette, W., and Bonow, J. T. (2009). e challenge of understanding process in
clinical behavior analysis: e case of functional analytic psychotherapy. Beha v.
Anal. 32, 135–148. doi: 10.1007/BF03392179
Forte, L. A., Timmer, S., and Urquiza, A. (2014). “A brief history of evidence-based
practice,” in Evidence-Based Approaches for the Treatment of Maltreated Children:
Considering Core Components and Treatment Eectiveness Child Maltreatment. eds.
S. Timmer and A. Urquiza (Dordrecht: Springer Netherlands), 13–18.
Froxán-Parga, M. X., Calero-Elvira, A., Pardo-Cebrián, R., and Núñez de
Prado-Gordillo, M. (2018). Verbal change and cognitive change: conceptual and
methodological analysis for the study of cognitive restructuring using the Socratic
dialog. J. Cogn. er. 11, 200–221. doi: 10.1007/s41811-018-0019-8
Froxán-Parga, M. X., Montserrat, M. F., Ana, C. E., and Elena, R. S. (2011).
Aproximación al Estudio Funcional de la Interacción Verbal entre Terapeuta y
Cliente Durante el Proceso Terapéutico. 1er. Premio de la XVII edición del Premio
de Psicología Aplicada “Rafael Burgaleta” 2010. Clínica y Salud 22, 69–85. doi:
10.5093/cl2011v22n1a5
Froxán-Parga, M. X., Ruiz-Sancho, E. M., and Calero-Elvira, A. (2016). A
theoretical and methodological proposal for the descriptive assessment of
therapeutic interactions. Psychother. Res. 26, 48–69. doi: 10.1080/
10503307.2014.935518
Hayes, S. C. (1981). Single case experimental design and empirical clinical
practice. J. Consult. Clin. Psychol. 49, 193–211. doi: 10.1037/0022-006X.49.2.193
Hayes, S. C. (2004). Acceptance and commitment therapy, relational frame theory,
and the third wave of behavioral and cognitive therapies. Behav. er. 35, 639–665.
doi: 10.1016/S0005-7894(04)80013-3
Hayes, S. C., and Hofmann, S. G. (2018). Process-based CBT: e Science and core
Clinical Competencies of Cognitive Behavioral erapy. Oakland, CA: New Harbinger
Publications, Inc.
Hayes, S. C., Strosahl, K. D., and Wilson, K. G. (1999). Acceptance and
Commitment erapy: An Experiential Approach to Behavior change. New York, NY,
US: Guilford Press.
Hedges, L. V., Pustejovsky, J. E., and Shadish, W. R. (2013). A standardized
mean dierence eect size for multiple baseline designs across individuals: eect
size for multiple baseline designs. Res. Synth. Methods 4, 324–341. doi: 10.1002/
jrsm.1086
Hofmann, S. G., Curtiss, J. E., and Hayes, S. C. (2020). Beyond linear mediation:
Toward a dynamic network approach to study treatment processes. Clin. Psychol.
Re v. 76:101824. doi: 10.1016/j.cpr.2020.101824
Horner, R. H., and Ferron, J. (2022). Advancing the application and use of single-
case research designs: reections on articles from the special issue. Perspect Behav.
Sci. 45, 5–12. doi: 10.1007/s40614-021-00322-x
Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., and Richman, G. S. (1994).
Toward a functional analysis of self-injury. J. Appl. Behav. Anal. 27, 197–209. doi:
10.1901/jaba.1994.27-197
Kazdin, A. E. (1978). History of Behavior Modication: Experimental Foundations
of Contemporary Research. Baltimore, MD: University Park Press.
Kazdin, A. E. (2021). Single-case experimental designs: characteristics, changes,
and challenges. J. Exp. Anal. Behav. 115, 56–85. doi: 10.1002/jeab.638
Kohlenberg, R. J., and Tsai, M. (1991). Functional analytic Psychotherapy: Creating
intense and Curative erapeutic Relationships. New York, NY, US: Plenum Press.
Kratochwill, T. R., and Levin, J. R. (2014). “Enhancing the scientic credibility of
single-case intervention research: Randomization to the rescue,” in Single-Case
Intervention Research: Methodological and Statistical Advances. School Psychology
Series. eds. T. R. Kratochwill and J. R. Levin (Washington, DC: American
Psychological Association), 53–89.
Lapresa, D., Arana, J., Anguera, M. T., and Garzón, B. (2013). Comparative
analysis of sequentiality using SDIS-GSEQ and THEME: A concrete example in
soccer. J. Sports Sci. 31, 1687–1695. doi: 10.1080/02640414.2013.796061
Lizarazo, N. E., Muñoz-Martínez, A. M., Santos, M. M., and Kanter, J. W. (2015).
A within-subjects evaluation of the eects of functional analytic psychotherapy on
in-session and out-of-session client behavior. Psychol. Rec. 65, 463–474. doi:
10.1007/s40732-015-0122-7
Magnús, S. M., Burgoon, J. K., and Casarrubea, M. (2016). Discovering hidden
temporal Patterns in Behavior and Interaction: t-Pattern Detection and Analysis with
THEMETM. New York, NY: Humana Press.
Magnusson, M. S. (2005). “Understanding social interaction: Discovering hidden
structure with model and algorithms,” in e Hidden Structure of Interaction: From
Neurons to Culture Patterns Emerging Communication: Studies in New Technologies
and Practices in Communication. eds. L. Anolli, S. Duncan and M. S. Magnússon
(Amsterdam, Netherlands: IOS Press), 3–22.
Magnusson, M. S. (2020). T-pattern detection and analysis (TPA) With THEMETM:
A mixed methods approach. Front. Psychol. 10:2663. doi: 10.3389/fpsyg.2019.02663
Manolov, R., Moeyaert, M., and Fingerhut, J. E. (2021). A priori justication for
eect measures in single-case experimental designs. Perspect Behav. Sci. 45, 153–186.
doi: 10.1007/s40614-021-00282-2
Mansueto, A. C., Wiers, R. W., van Weert, J. C. M., Schouten, B. C., and
Epskamp, S. (2022). Investigating the feasibility of idiographic network models.
Psychol. Methods. doi: 10.1037/met0000466 [Epub Ahead of Print]
McHugh, M. L. (2012). Interrater reliability: the kappa statistic. Biochem. Med. 22,
276–282. doi: 10.11613/BM.2012.031
Molenaar, P. C. (2004). A manifesto on psychology as idiographic science:
bringing the person back into scientic psychology, t his time forever. Measurement
2, 201–218. doi: 10.1207/s15366359mea0204_1
Moore, J. (1991). A retrospective appreciation of Willard Day’s contributions to
radical behaviorism and the analysis of verbal behavior. Analysis of Verbal Behav. 9,
97–104. doi: 10.1007/BF03392863
Pardo-Cebrian, R., Virues-Ortega, J., Calero-Elvira, A., and Guerrero-Escagedo, M. C.
(2021). Toward an experimental analysis of verbal shaping in psychotherapy. Psychother.
Res. 32, 497–510. doi: 10.1080/10503307.2021.1955418
Parker, R. I., and Vannest, K. (2009). An improved eect size for single-case research:
nonoverlap of All pairs. Behav. er. 40, 357–367. doi: 10.1016/j.beth.2008.10.006
Pattern Vision (2021). emeEdu. Reykjavik, Iceland: Pattern Vision. Available
at: http://patternvision.com/ (Accessed March 6, 2021).
Paul, G. L. (1969). “Behavior modication research: design and tactics,” in Behavior
erapy: Appraisal and status. ed. C. M. Franks (Ne w York: McGraw-Hill), 29–62.
Piccirillo, M. L., Beck, E. D., and Rodebaugh, T. L. (2019). A clinician’s primer for
idiographic research: considerations and recommendations. Behav. er. 50,
938–951. doi: 10.1016/j.beth.2019.02.002
Pustejovsky, J. E. (2018). Using response ratios for meta-analyzing single-case designs
with behavioral outcomes. J. Sch. Psychol. 68, 99–112. doi: 10.1016/j.jsp.2018.02.003
Santoyo, C., Jonsson, G. K., Anguera, M. T., and López-López, J. A. (2017).
Observational analysis of the organization of on-task behavior in the classroom
using complementary data analysis. Anales de Psicología 33, 497–514. doi: 10.6018/
analesps.33.3.271061
Tate, R. L., Perdices, M., Rosenkoetter, U., Shadish, W., Vohra, S., Barlow, D. H.,
et al. (2016). e single-case reporting guideline In BEhavioural interventions
(SCRIBE) 2016 statement. Phys. er. 96, e1–e10. doi: 10.2522/ptj.2016.96.7.e1
Tolin, D. F., McKay, D., Forman, E. M., Klonsky, E. D., and ombs, B. D. (2015).
Empirically supported treatment: recommendations for a new model. Clin. Psychol.
Sci. Pract. 22, 317–338. doi: 10.1111/cpsp.12122
Tsai, M., Yard, S. S., and Kohlenberg, R. (2014). Functional analytic psychotherapy:
a behavioral relational approach to treatment. Psychotherapy 51, 364–371. doi:
10.1037/a0036506
Van Agteren, J., Iasiello, M., Lo, L., Bartholomaeus, J., Kopsais, Z., Carey, M.,
et al. (2021). A systematic review and meta-analysis of psychological interventions
to improve mental wellbeing. Nat. Hum. Behav. 5, 631–652. doi: 10.1038/
s41562-021-01093-w
Alonso-Vega et al. 10.3389/fpsyg.2022.949733
Frontiers in Psychology 11 frontiersin.org
Virues-Ortega, J., and Froxán-Parga, M. X. (2015). A translational approach to the
functional analysis of language in psychotherapy. Int. J. Clin. Health Psychol. 15,
69–75. doi: 10.1016/j.ijchp.2014.10.001
Wright, A. G. C., and Woods, W. C. (2020). Personalized models of
psychopathology. Annu. Rev. Clin. Psychol. 16, 49–74. doi: 10.1146/annurev-
clinpsy-102419-125032
Xavier, R. N., Kanter, J. W., and Meyer, S. B. (2012). Transitional probability
analysis of two child behavior analytic therapy cases. Int. J. Behav. Consult. er. 7,
182–188. doi: 10.1037/h0100954
Zamignani, D. R. (2008). O desenvolvimento de um sistema multidimensional
para a categorização de comportamentos na interação terapêutica. Doctoral
dissertation. Brazil: Universidade de São Paulo.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Our current understanding of the efficacy of psychological interventions in improving mental states of wellbeing is incomplete. This study aimed to overcome limitations of previous reviews by examining the efficacy of distinct types of psychological interventions, irrespective of their theoretical underpinning, and the impact of various moderators, in a unified systematic review and meta-analysis. Four-hundred-and-nineteen randomized controlled trials from clinical and non-clinical populations (n = 53,288) were identified for inclusion. Mindfulness-based and multi-component positive psychological interventions demonstrated the greatest efficacy in both clinical and non-clinical populations. Meta-analyses also found that singular positive psychological interventions, cognitive and behavioural therapy-based, acceptance and commitment therapy-based, and reminiscence interventions were impactful. Effect sizes were moderate at best, but differed according to target population and moderator, most notably intervention intensity. The evidence quality was generally low to moderate. While the evidence requires further advancement, the review provides insight into how psychological interventions can be designed to improve mental wellbeing. This meta-analysis of 419 randomized controlled trials found that various types of psychological interventions could improve mental wellbeing in clinical and non-clinical populations. Effect sizes tended to be small to moderate and were influenced by various moderators.
Article
Full-text available
Due to the complex nature of single-case experimental design data, numerous effect measures are available to quantify and evaluate the effectiveness of an intervention. An inappropriate choice of the effect measure can result in a misrepresentation of the intervention effectiveness and this can have far-reaching implications for theory, practice, and policymaking. As guidelines for reporting appropriate justification for selecting an effect measure are missing, the first aim is to identify the relevant dimensions for effect measure selection and justification prior to data gathering. The second aim is to use these dimensions to construct a user-friendly flowchart or decision tree guiding applied researchers in this process. The use of the flowchart is illustrated in the context of a preregistered protocol. This is the first study that attempts to propose reporting guidelines to justify the effect measure choice, before collecting the data, to avoid selective reporting of the largest quantifications of an effect. A proper justification, less prone to confirmation bias, and transparent and explicit reporting can enhance the credibility of the single-case design study findings.
Article
Full-text available
Abstract: The aim of this study was to analyze the organization of on-task behavior in the classroom. Four observational methodology techniques— T-pattern detection, lag sequential analysis, trend analysis, and polar coordinate analysis—were used to study the organization of on-task and offtask behavioral patterns during class time in a primary school setting. The specific objective was to detect and explore relationships between on-task behavior and different social interaction categories in relation to the actual distribution of activities in a real-life classroom setting. The study was conducted using the behavioral observation system for social interaction SOCIS and the software programs Theme (version 6, Edu), SDIS-GSEQ (version 4.1.2), HOISAN (version 1.6), and STATGRAPHICS (version 6). We describe the results obtained for the four techniques and discuss the methodological implications of combining complementary techniques in a single study. Key words: Observational methodology; social interaction; academic engagement; T-pattern; lag sequential analysis; trend analysis; polar coordinates.
Preprint
Psychological interventions have shown efficacy for the treatment of different psychological problems. Now, we need to know what processes make these results possible. The observational analysis of the verbal interaction of the therapist and the client has been used to analyze the learning processes that underlie the change in the client's behavior. However, previous coding systems have conceptual and methodological limitations. Therefore, the purpose of this article is to develop a functional coding system for the analysis of verbal interaction in clinical contexts. To this end, we developed and debugged a new functional description-based system. Also, we have analyzed 66 clinical sessions belonging to 23 different cases using the new coding system and the results have been compared with previous studies. The main result of this study is a system consisting of 5 functional categories with good reliability indices and methodological properties. The discriminant and cluster analysis have yielded similar results to those of previous studies. Finally, we discuss about conceptual and methodological advances of this new system, we list the main limitations, and we propose further studies in this line of research.
Article
Recent times have seen a call for personalized psychotherapy and tailored communication during treatment, leading to the necessity to model the complex dynamics of mental disorders in a single subject. To this aim, time-series data in one patient can be collected through ecological momentary assessment and analyzed with the graphical vector autoregressive model, estimating temporal and contemporaneous idiographic networks. Idiographic networks graph interindividual processes that may be potentially used to tailor psychotherapy and provide personalized feedback to clients and are regarded as a promising tool for clinical practice. However, the question whether we can reliably estimate them in clinical settings remains unanswered. We conducted a large-scale simulation study in the context of psychopathology, testing the performance of personalized networks with different numbers of time points, percentages of missing data, and estimation methods. Results indicate that sensitivity is low with sample sizes feasible for clinical practice (75 and 100 time points). It seems possible to retrieve the global network structure but not to recover the full network. Estimating temporal networks appears particularly challenging; thus, with 75 and 100 observations, it is advisable to reduce the number of nodes to around six variables. With regard to missing data, full information maximum likelihood and the Kalman filter are effective in addressing random item-level missing data; consequently, planned missingness is a valid method to deal with missing data. We discuss possible methodological and clinical solutions to the challenges raised in this work. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
Article
This special issue of Perspective on Behavior Science is a productive contribution to current advances in the use and documentation of single-case research designs. We focus in this article on major themes emphasized by the articles in this issue and suggest directions for improving professional standards focused on the design, analysis, and dissemination of single-case research.
Article
Objective. Some authors have suggested that verbal shaping is an active component in cognitive restructuring. The goal of the current study was to provide experimental evidence in support of the shaping hypothesis. Method. Three consecutively admitted adults with symptoms of anxiety and depression participated in the study. Participants underwent baseline, treatment comparison, crossover, and follow-up phases as part of an experimental single-case study. We selected two target irrational beliefs and praised either the client’s approximations to a terminal (more adaptive) class of verbalizations (differential reinforcement of approximations) or just the terminal class of verbalizations (terminal reinforcement). Results. Irrational beliefs exposed to differential reinforcement of approximations was a more effective approach to reducing irrational beliefs. Conclusion. The study provides experimental evidence in support of the verbal shaping hypothesis. Our findings suggest that it may be possible to enhance clinical outcomes by actively engaging in verbal shaping during client-therapist verbal interaction.
Article
Tactics of Scientific Research (Sidman, 1960) provides a visionary treatise on single‐case designs, their scientific underpinnings, and their critical role in understanding behavior. Since the foundational base was provided, single‐case designs have proliferated especially in areas of application where they have been used to evaluate interventions with an extraordinary range of clients, settings, and target foci. This article highlights core features of single‐case experimental designs, how key and ancillary features of the designs have evolved, the special strengths of the designs, and challenges that have impeded their integration in many areas where their contributions are sorely needed. The article ends by placing the methodological approach in the context of other research traditions. In this way, the discussion moves from the specific designs toward foundations and philosophy of science issues in keeping with the strengths of the person and book we are honoring.
Article
The personalized approach to psychopathology conceptualizes mental disorder as a complex system of contextualized dynamic processes that is nontrivially specific to each individual, and it seeks to develop formal idiographic statistical models to represent these individual processes. Although the personalized approach draws on long-standing influences in clinical psychology, there has been an explosion of research in recent years following the development of intensive longitudinal data capture and statistical techniques that facilitate modeling of the dynamic processes of each individual's pathology. Advances are also making idiographic analyses scalable and generalizable. We review emerging research using the personalized approach in descriptive psychopathology, precision assessment, and treatment selection and tailoring, and we identify future challenges and areas in need of additional research. The personalized approach to psychopathology holds promise to resolve thorny diagnostic issues, generate novel insights, and improve the timing and efficacy of interventions. Expected final online publication date for the Annual Review of Clinical Psychology, Volume 16 is May 7, 2020. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.