Supportive Psychotherapy Versus Interpersonal Psychotherapy

Functional Analytic Psychotherapy Compared With Supportive Listening: An Alternating Treatments Design Examining

Distinctiveness, Session Evaluations, and Interpersonal Functioning

Daniel W. M. Maitland and Scott T. Gaynor Western Michigan University

Functional analytic psychotherapy (FAP) is based on the notion that because the therapeutic relationship is an interpersonal one, it will share some functional similar- ities with other relationships. When client responses that are problematic or adaptive in daily life occur in the therapy session, the therapist can provide immediate response contingent consequences. FAP’s in-session focus has been hypothesized to produce particularly intense and effective therapy sessions and therapeutic relationships that are especially meaningful. These hypotheses were tested in a restricted alternating treat- ments design offering 13 participants (M age � 19.54 years, 77% female) reporting difficulties in interpersonal relating sessions of FAP or supportive listening (SL). Eighty (41 SL and 39 FAP) of 107 (56 SL and 51 FAP) sessions were coded for the elements of FAP and SL. FAP sessions were marked by a statistically significant difference in the application of the in-session focus distinctive of FAP. Both treatments were high in client-centered interaction; however, the results favored SL. Consistent small to moderate effects favoring FAP sessions were observed on the Working Alliance Inventory – Short Form (WAI-SF; Busseri & Tyler, 2003), Session Rating Scale (SRS; Duncan et al., 2003), and FAP Session Bridging Form (FSBF; Tsai et al., 2009). These data are consistent with FAP-generated hypotheses: In-session contingent responding contributes to more engaging sessions and stronger therapeutic alliances. The findings have implications for FAP efficacy research, particularly the suggestion that the therapeutic alliance may serve as a proxy for FAP’s mechanism of action.

Keywords: functional analytic psychotherapy, therapeutic relationship, interpersonal functioning

There is evidence for the effectiveness of psychotherapy (American Psychological Asso- ciation, 2012), but fundamental questions re- main. Meta-analytic data suggest the therapeu- tic alliance is associated with positive outcomes (Horvath, 2011; Martin, Garske, & Davis, 2000; Orlinsky, Rønnestad, & Willutzki, 2003); how- ever, the precise nature and direction of the relationship remains unclear (Kazdin, 2007; Siev, Huppert, & Chambless, 2009), such that

the alliance is often referred to as a nonspecific or common factor (R. P. Greenberg, 2012; Ilardi & Craighead, 1994). How is it that the relation- ship between a therapist and client can be cu- rative? Functional analytic psychotherapy (FAP; Kanter, Tsai, & Kohlenberg, 2010; Kohlenberg & Tsai, 1991; Tsai et al., 2009) provides a behavioral interpretation of how the interactions occurring in the therapeutic rela- tionship can produce enhanced client outcomes. The core idea is that because the therapeutic relationship is an interpersonal one, it will share some functional similarities with other relation- ships. Consequently, the behaviors that are problematic (and adaptive) for the client in his or her daily interactions are expected to occur with the therapist. When these interpersonal re- sponse classes occur, the therapist is in a posi- tion to provide response-contingent conse- quences in the session to decrease problematic

Daniel W. M. Maitland and Scott T. Gaynor, Department of Psychology, Western Michigan University.

Daniel W. M. Maitland is now at Center for the Science of Social Connection, University of Washington.

Correspondence concerning this article should be ad- dressed to Daniel W. M. Maitland, Center for the Science of Social Connection, Department of Psychology, University of Washington, Seattle, WA 98195. E-mail: danmaitland@ gmail.com

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Behavior Analysis: Research and Practice © 2016 American Psychological Association 2016, Vol. 16, No. 2, 52– 64 2372-9414/16/$12.00 http://dx.doi.org/10.1037/bar0000037

52

response classes and increase adaptive ones. The consequences provided are to be those that are natural or inherent to interpersonal interac- tion, such that the client should be more likely to emit this behavior in his or her daily life in a fitting interpersonal context. Thus, the mecha- nism of action in FAP is operant conditioning, namely, therapist-provided consequences con- tingent on client in-session responding (Kanter et al., 2010; Kohlenberg & Tsai, 1991; Tsai et al., 2009).

To promote implementation, FAP identifies three client response classes upon which thera- pists should focus. These are termed clinically relevant behaviors (CRBs). CRB1s are collab- oratively identified problem behaviors that oc- cur in session. CRB2s are improvements that occur in session, and CRB3s are attempts by the client to describe the contingencies controlling his or her behavior. The therapist’s job is to respond to instances of CRBs in a manner that reduces CRB1s, and prompts, shapes, and in- creases CRB2s and CRB3s. The therapist is guided in the task of responding to CRBs, and thereby implementing the mechanism of action, by five rules: (a) Watch for CRBs, (b) Evoke CRBs, (c) Reinforce CRBs, (d) Observe the potentially reinforcing effects of therapist be- havior in relation to suspected client CRBs, and (e) Provide functional analytically informed in- terpretations and implement generalization strategies, including assignment of relevant homework that parallels in session responding (Kohlenberg & Tsai, 1991; Tsai et al., 2009; Weeks, Kanter, Bonow, Landes, & Busch, 2012). Observing, evoking, and reinforcing CRBs requires identifying CRBs. Identifying CRBs is easier said than done because CRBs are idiographic—a CRB1 for one client might be a CRB2 for another. CRBs must be determined individually based on their function for a par- ticular person. Although some general guide- lines for evoking and reinforcing CRBs can be offered, the therapist behaviors that will func- tion as discriminative and reinforcing stimuli for CRB will be, at least to some extent, idio- syncratic (Follette & Bonow, 2009). This func- tional idiographic nature of FAP has made it challenging to specify as an independent vari- able for efficacy research (Maitland & Gaynor, 2012).

Maitland and Gaynor (2012) described the rationale for a line of FAP efficacy research

informed by influential treatment development methods in empirical clinical psychology (Ro- unsaville, Carroll, & Onken, 2001) and contem- porary behavior therapies (Vilardaga, Hayes, Levin, & Muto, 2009). An early step involves determining whether FAP could be distin- guished as an independent variable. That is, could an adherence coder differentiate FAP ses- sions from a reasonable alternative? Supportive listening (SL) was offered as a relevant and important early comparison, as it shares with FAP an emphasis on establishing a strong ther- apeutic relationship marked by genuine positive regard and empathic attunement (Rogers, 1957), but should differ on the critical variable unique to FAP—in-session contingent respond- ing (Maitland & Gaynor, 2012).

In order for therapist-contingent responses to reduce CRB1s and prompt, shape, and increase CRB2s and CRB3s, therapist responses must have consequential (e.g., reinforcing, punish- ing) functions. In the absence of established methods for formally assessing and comparing reinforcers in outpatient psychotherapy, a rea- sonable proxy (for predicting that therapist re- sponding could have consequential functions) is a therapeutic relationship marked by a strong bond and a collaborative agreement on the goals and tasks of therapy. That is, a therapeutic alli- ance (Bordin, 1979) appears to be an important prerequisite for engagement of FAP’s mecha- nism of action (Maitland & Gaynor, 2012).

Flexible application of the five FAP rules guides the therapist in implementing the mech- anism of action. In discussing only Rule 1, Kohlenberg and Tsai (1991, p. 24; see also Kohlenberg & Tsai, 1994) indicated,

Our major hypothesis is that following this rule im- proves therapeutic outcome. . . . It is also hypothesized that following Rule 1 will lead to increased intensity; that is, stronger emotional reactions of the therapist and client to each other during the session.

Logic suggests that such an effect should only be further enhanced by application of the full range of FAP guidelines. FAP’s strong here-and-now, in-session focus should lead to particularly intense and engaging therapy ses- sions, and therapeutic relationships that are meaningful, intimate, and curative (Kohlenberg & Tsai, 1991; Tsai et al., 2009). As such, FAP sessions should be rated more positively than

53ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

those based simply on provision of empathic understanding and positive regard (i.e., SL).

The current study used a restricted alternating treatments design to compare FAP sessions against SL sessions. Participants with self- reported difficulties in social relating received six to 10 weekly therapy sessions. The sequenc- ing of sessions (SL or FAP) was determined at random, with the restriction that the first session was SL and the last session FAP, and that no more than two consecutive sessions of the same treatment could occur in a row. The following hypotheses were examined:

1. FAP sessions will differ from SL sessions in the amount of in-session contingent therapist responding (i.e., the application of FAP rules).

2. Session evaluations will be higher follow- ing FAP than SL sessions.

3. Participants will report improved daily life interpersonal relating with changes con- centrated in weeks following FAP ses- sions compared with SL sessions.

Method

Participants

University students were recruited into an intervention study for individuals reporting so- cial difficulties. Inclusion required scoring one standard deviation below the gender-based nor- mative mean on the Fear of Intimacy Scale (FIS; Descutner & Thelen, 1991) and the Miller Social Intimacy Scale (MSIS; R. S. Miller, & Lefcourt, 1982). Participants were 18 or older, fluent English speakers, not receiving psycho- therapy, and, if taking psychotropic medica- tions, at the current dose for at least 6 months. Exclusion criteria consisted of meeting diagnos- tic criteria for posttraumatic stress disorder; a history of psychosis, substance dependence, or obsessive– compulsive disorder; or presenting as an imminent suicide risk.

Nineteen potential participants were screened. Of those, 14 met inclusion criteria; none were excluded. One participant discontinued after the assessment session, yielding 13 participants who engaged in at least one treatment session and were included in analyses. These 13 full-time collegians had a mean age of 19.54 years (SD � 2.50). Ten (77%) were female, five (39%) were freshmen,

five (39%) were sophomores, one (8%) was a junior, and two (15%) were seniors. Eight (61%) reported being Caucasian, three (23%) reported being African American, one (8%) reported being Asian American, and one (8%) reported being multiracial. Five (39%) reported a history of coun- seling (for depression, anxiety, anorexia or family issues), and three (23%) had a history of taking selective serotonin reuptake inhibitors. As re- quired for inclusion, participants reported im- paired social-relating on the FIS (M � 118.08, SD � 13.35) and MSIS (M � 108.08, SD � 11.31).

Materials

Postsession evaluations. Three measures were collected after each session. The 12-item Working Alliance Inventory – Short Form (WAI-SF; Busseri & Tyler, 2003) has shown strong internal consistency (� � .92; Hanson, Curry, & Bandalos, 2002) and correlates highly with the WAI (Horvath & Greenberg, 1989). The WAI-SF measures aspects of the therapeu- tic alliance considered common to all treat- ments. Higher scores (range 12 to 84) indicate a better alliance. The Session Rating Scale (SRS; Duncan et al., 2003) is generally used as a clinical tool, but has been examined psychomet- rically with a clinical sample (� � .88, r � .64 with Helping Alliance Questionnaire). The four-item SRS uses a 100-mm visual analog scale to indicate measure clients’ evaluation of the relationship, goals and topics, approach or method, and an overall session rating. The WAI-SF and SRS come from the common fac- tors tradition. The FAP Session Bridging Form (FSBF; Tsai et al., 2009) is a clinical tool from the FAP literature. The FSBF has four quanti- tative items scored on 1–10 scale reflecting the participants’ report of the helpfulness/effective- ness of the session, and his or her feelings of connectedness, engagement, and presence in the session. The total of the four FSBF quantitative items was analyzed.

Measures of interpersonal functioning. Changes in daily life interpersonal relating were measured using the FIS (Descutner & Thelen, 1991) and the MSIS (R. S. Miller, & Lefcourt, 1982). FIS and MSIS data collected prior to treatment and after the conclusion of treatment were available from 10 of 13 participants. Last observation carried forward was used when

54 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

posttreatment were missing, which, in this cir- cumstance, assumes no change. The FIS is a 35-item measure of anxiety about self-disclo- sure and social intimacy in close relationships. Normative data suggest a mean of 78.75 (men � 81.9, women � 76.10), with a standard deviation of 21.82 (men � 20.58, women � 22.61), � � .93, and a 1-month test–retest reli- ability of r � .89. The MSIS is a 17-item questionnaire assessing behavior in a meaning- ful relationship. Normative data for unmarried collegians was 137.5 (males � 134.9, fe- males � 139.3), with a standard deviation of 19.1 (males � 21.9, females � 16.8), � � .91, and a 1-month test–retest reliability of r � .96.

Weekly changes were measured using the Outcome Rating Scale (ORS; S. D. Miller, Dun- can, Brown, Sparks, & Claud, 2003) adminis- tered prior to each therapy session. The four- item ORS requires participants mark on a 100-mm visual analog scale how they have been feeling in four domains (overall, individually, interpersonally, and socially). S. D. Miller et al. (2003) reported high internal consistency, � � .93, and test–retest reliability of .66, .58, and .49, at second, third, and fourth administrations. A significant correlation, r � .59, was found between the ORS total score and the Outcome Questionnaire-45 (Lambert et al., 1996).

Procedure

Following informed consent, two 1-hr assess- ments occurred, which involved filling out questionnaires, completing a clinical interview, and collecting background information. In- cluded participants were offered 10 sessions (five SL and five FAP). The first was always SL and the last was FAP. The remaining sessions were randomized, in blocks of two, such that no more than two SL or FAP sessions could occur in a row and that attendance at all 10 would result in receipt of five each of SL and FAP. Thus, the methodology was an alternating treat- ments design. The 13 study participants re- ceived a total of 107 sessions (M � 8.23, SD � 1.64, range 5 to 10), 56 SL (M � 4.31, SD � 0.75, range 3 to 5) and 51 FAP (M � 3.92, SD � 0.95, range 2 to 5).

Treatment conditions. FAP condition. FAP sessions attempted

to represent what Maitland and Gaynor (2012) described as a Level 3 application of

FAP, distinguished by an attempt to follow all five FAP rules (see also Kohlenberg, 2005). FAP focused on using therapist responses as a reinforcer to shape interpersonal behavior. This occurred when the therapist contingently responded to collaboratively identified CRBs. To facilitate contingent responding, there was an attempt during the initial assessment phase to establish a working therapeutic alliance, which included a descriptive introduction to FAP (see Tsai et al., 2009, pp. 71–73), dis- cussion of how it might work for social relat- ing, and a collaborative conceptualizing of CRBs. Establishing a general early agreement on goals and tasks allowed the majority of FAP sessions to be akin to those described in the middle phase or end phase in Tsai et al. (2009). That is, the therapist attempted to actively implement the mechanism of action– contingent responding to CRB.

In addition, the therapist looked for oppor- tunities to complete a logical FAP interaction (Weeks et al., 2012). A logical FAP interac- tion begins with an out-to-in parallel based on therapist observation of a similarity between an outside problem and a CRB1, which is then used to evoke and reinforce CRB2s. The ther- apist then checks to see if his or her response was reinforcing, attempts to generalize, and then provides an in-to-out parallel. Special attention was also paid to the ending of the therapeutic relationship as this provided an opportunity to explore CRB1s and CRB2s related to the ending of relationships.

SL condition. The SL condition focused on becoming more aware of daily life emo- tional experience and interpersonal relation- ships without actively attempting to respond to in-session behavior or to draw explicit par- allels to in-session behavior. SL sessions al- lowed unpunished expression of daily-life re- lational experiences. The therapist focused on asking open-ended questions, offering em- pathic and reflective responses, communicat- ing and expressing empathy and interest, and asking clarifying questions. A protocol that has been used previously (see Clore & Gaynor, 2012) served as the basis for the SL sessions.

Treatment integrity. The first author was the therapist. He had completed over a year of doctoral clinical coursework and practicum. FAP specific training included reading A

55ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Guide to Functional Analytic Psychotherapy (Tsai et al., 2009) and participating in nine FAP workshops, including advanced and in- tensive trainings. In regard to SL training, the therapist read and discussed L. S. Greenberg, Watson, and Lietaer (1998), read and dis- cussed a partial transcript of Carl Rogers (from Corsini & Wedding, 2005), and watched and discussed Rogers’s interview with “Gloria” (Psychological & Educational Films, 1981). The therapist then reviewed and practiced conducting sessions using an SL protocol (from Clore & Gaynor, 2012). Ther- apist preparation and treatment delivery dur- ing the study was supervised by the second author, who has a long familiarity with FAP and considerable experience conducting clin- ical outcome research, including outcome re- search with FAP (e.g., Gaynor & Lawrence, 2002).

To confirm that the intended type of ther- apy was delivered, 100% of available FAP and SL sessions were coded for adherence. Sessions from one participant and some early sessions for several other participants were not available because of audio system failure. For the remaining 12 participants, four to nine sessions (M � 6.67, SD � 1.83) were coded for each. In total, 41 SL sessions and 39 FAP sessions (80/107 � 75% of the total sessions) were able to be coded for the presence of the SL elements and application of the FAP rules. The coder was a doctoral student in clinical psychology with knowledge of FAP from graduate coursework and attending a 2-day workshop. The coder was trained on the ad- herence measure, which required some a pri- ori knowledge of what to be looking for while coding, but was kept unaware of session type at the time of coding.

The adherence measure included 10 items—four related to SL (i.e., the SL sub- scale: frequency of attempting to understand the daily life social relationships from the client’s vantage point, frequency of engaging in reflective and empathic listening, fre- quency of prompts to discuss daily life social relations, and frequency of turning the focus to the client’s feelings/emotional reactions to events in their daily life) and five related to the FAP rules (i.e., the FAP subscale: fre- quency of turning the focus to in-session be- havior, frequency of comparing in-session

events with the participant’s daily life, fre- quency of prompts to engage in a particular response in session, frequency of the therapist sharing his or her reaction to the client’s behavior, and frequency of checking to see the client’s reaction to the therapist sharing his or her reaction). Items 1 to 9 were scored 0 (did not occur), 1 (occurred once), 2 (oc- curred twice), or 3 (occurred three or more times). The 10th item assessed assignment of homework, which was scored 0 (did not oc- cur), 1 (partial), or 2 (occurred). In SL ses- sions, homework involved emotion monitor- ing. FAP homework was to follow from in session events. The result was that partial or complete homework assignment was ob- served in 39/41 (95% of) SL sessions, but only 20/39 (51% of) FAP sessions.

Data analysis. Hypothesis 1 was tested with paired samples t tests using the SL and FAP subscale mean scores from the adherence coding. For Hypothesis 2, the initial analytic strategy involved visual inspection of plotted data along with conducting randomization tests (RT; Heyvaert & Onghena, 2014) and calculat- ing individual nonoverlap of all pairs (NAP) effect sizes (Parker & Vannest, 2009). How- ever, it became apparent that use of RT and NAP statistics were complicated by two aspects of the data: (a) carryover effects (the alliance tended to improve over the course of treatment for some participants), and (b) ceiling effects (the alliance was often rated very highly, near the top of the range). Ceiling effects shrink the potential range such that the magnitude of dif- ferences that might be obtained between the treatments was diminished, thus weakening RTs. A trend upward increases the overlap be- tween data points in the separate conditions weakening the nonoverlap analyses. To illus- trate, the data for all participants on the WAI-SF are presented in Figure 1. Part of the reason for the carryover and ceiling effects may be that both treatments contain SL components, which are widely thought to contribute to the thera- peutic bond. Given this overlap, and the visual and statistical examination of the data in Figure 1, strategies were needed that might reveal small to moderate but consistent differences. As such, several additional analytic approaches were used.

First, means were calculated for each par- ticipant for each condition and compared us-

56 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

ing paired t tests and within-group effect size estimates (g). Second, the proportion of mean session rating scores that were higher for FAP than SL was tested with binomial sign tests.

Third, weighted average NAP scores across the sample was tested for statistical signifi- cance (Vannest, Parker, & Gonen, 2011). Fourth, conditional probabilities were exam-

12

24

36

48

60

72

84

P2

12

24

36

48

60

72

84

P6

12

24

36

48

60

72

84

P7

12

24

36

48

60

72

84

P8

12

24

36

48

60

72

84

P9

12

24

36

48

60

72

84

P10

12

24

36

48

60

72

84

P11

12

24

36

48

60

72

84

P18

RT B-A = 5.00, p = .06 NAP = .86, p = .06 PEM = 100%

RT B-A = 1.33, p = .26 NAP = .66, p = .47 PEM = 50%

RT B-A = 0.75, p = .50 NAP = .64, p = .46 PEM = 0%

RT B-A = 2.42, p = .22 NAP = .72. p = .31 PEM = 75%

RT B-A = 2.67, p = .07 NAP = .78, p = .18 PEM = 67%

RT B-A = 7.00, p = .07 NAP = 1.00, p = .03 PEM = 1.00

RT B-A = 1.25, p = .27 NAP = .76, p = .17 PEM = 100%

RT B-A = 0.67, p .43 NAP = .67, p = .48 PEM = 33%

Figure 1. Working Alliance Inventory – Short Form scores after each session of functional analytic psychotherapy and supportive listening along with the mean for each condition for each participant. Each panel also shows the results of randomization test (RT), nonoverlap of all pairs (NAP), and percentage of data points exceeding median (PEM) analyses for each participant.

57ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

ined—that is, the probability of a higher ses- sion rating after a FAP session preceded by an SL session, and vice versa. Odds ratios were then calculated to test whether FAP sessions following SL sessions led to higher session ratings. For Hypothesis 3, paired samples t tests compared the mean ORS score in the weeks following both FAP and SL sessions with that from the week preceding the first therapy session and each other.

Results

Hypothesis 1: FAP Sessions Will Differ From SL Sessions in the Application of FAP Rules

If the treatments were not distinguishable in this way, there would be no basis for expecting

differential postsession ratings. The adherence coding data resulted in a mean (SD) on the SL subscale for the 41 SL sessions of 2.98 (0.07), whereas the mean (SD) on the FAP subscale for the SL sessions was 0.15 (0.41). These data suggest SL sessions were marked by the clear presence of client-centered engagement, but not in-session contingent responding characteristic of FAP. The mean (SD) from the 39 FAP ses- sions was 2.73 (0.41) on the SL subscale and 2.36 (0.51) on the FAP subscale. These FAP session data suggest that although FAP sessions were also marked by the presence of client- centered engagement, they also included the distinctive FAP elements.

The average session rating on both the SL subscale and FAP subscale was calculated for each participant for each therapy type allowing for paired samples t tests to be conducted on the

12

24

36

48

60

72

84

P25

12

24

36

48

60

72

84

P26

12

24

36

48

60

72

84

P27

12

24

36

48

60

72

84

P28

12

24

36

48

60

72

84

P29

RT B-A = 8.50, p < .01 NAP = 1.00, p = .05 PEM = 100%

RT B-A = 2.33, p = .43 NAP = .75, p = .25 PEM = 67%

RT B-A = -1.25, p = .76 NAP = .48, p = .92 PEM = 0%

RT B-A = 2.50, p = .21 NAP = .80, p = .14 PEM = 75%

RT B-A = -6.00, p = 1.0 NAP = .25, p = .61 PEM = 0%

Figure 1. (continued)

58 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

SL subscale averages and the FAP subscale averages. The SL subscale average in SL ses- sions (M � 2.98, SD � 0.04) and FAP sessions (M � 2.74, SD � 0.24) was significantly dif- ferent, t(11) � 3.81, p � .003, suggesting higher SL scores in SL sessions. The FAP sub- scale average in SL sessions (M � 0.15, SD � 0.17) and FAP sessions (M � 2.35, SD � 0.39) was highly statistically significant, t(11) � 20. 97, p � .001, suggesting higher FAP scores in FAP sessions. These data suggest that although both SL and FAP contained supportive ele- ments, they were quite different in terms of their focus on daily life versus in-session contingent responding, the latter being the distinctive com- ponent of FAP.

Hypothesis 2: Postsession Evaluations Will Be Higher for FAP Than SL Sessions

Given that the sessions were clearly unique in their application of the FAP rules, it is possible to assess whether the FAP sessions were more positively evaluated. The WAI-SF session (and overall mean) scores are presented in Figure 1, which also includes the individual RT and NAP analyses. The generally high WAI-SF scores (ceiling effect) and cases of increasing general trends (carryover effect) are apparent in visual inspection of the figure. Also apparent is that only a small number of RT and NAP analyses were statistically significant. According to ten- tative interpretive ranges for NAP analyses (0% to 65% � weak, 66% to 92% � moderate, and 93% to 100% � strong; Parker & Vannest, 2009) effects were strong for two participants, moderate for eight participants, and weak for three participants. The weighted NAP analysis suggested a statistically significant, moderate effect size advantage for FAP, Z � 5.92, p � .001, NAP � .72. Likewise, the mean WAI-SF score was higher for the FAP sessions than SL sessions for 11/13 (85%), yielding a significant binomial sign test favoring FAP, p � .03. More- over, although the WAI-SF revealed high means for both conditions, there was a statisti- cally significant, small effect size advantage for FAP (M � 77.46, SD � 8.89) over SL (M � 74.64, SD � 10.05), t(12) � �2.71, p � .02, g � .28. The weight of the WAI-SF data sug- gests a robust small to moderate effect favoring FAP.

The probability of a higher WAI-SF after a FAP session preceded by an SL session was 23/40 � .58, and the odds 23/17 � 1.35. The probability of a higher WAI-SF after an SL session preceded by a FAP session was 7/28 � .25, and the odds 7/21 � .33. The result is a significant odds ratio of 1.35/.33 � 4.06 (95% confidence interval [CI] [1.41, 11.72]), Fisher’s Exact Test (FET) p � .01, favoring FAP. The probability of a higher WAI-SF after an SL session preceded by an SL session was 7/15 � .47, and the odds 7/8 � .88. The probability of a higher WAI-SF after a FAP session preceded by a FAP session was 4/9 � .44, and the odds 4/5 � .80. The result is a nonsignificant odds ratio of .88/.80 � 1.09 (95% CI [0.21, 5.76]), FET p � 1.00. Increases on the WAI-SF were most likely to occur when a FAP session fol- lowed an SL session.

On the SRS, the NAP effect sizes were mod- erate for 9 participants and weak for 4 partici- pants. The group-wise NAP analysis suggested a statistically significant advantage for FAP, Z � 5.597, p � .001, with a moderate effect size, NAP � .69. The mean SRS score was higher for the FAP sessions than SL for 11/13 (85%), yielding a statistically significant bino- mial sign test favoring FAP, p � .03. High means were again seen in both treatments, but with a trend toward FAP (M � 95.85, SD � 4.81) being higher than SL (M � 94.15, SD � 4.36), t(12) � �1.85 p � .09, g � .35. The SRS data as a whole suggest a small-moderate effect favoring FAP.

The probability of a higher SRS after a FAP session preceded by an SL session was 27/40 � .68, and the odds 27/13 � 2.08. The probability of a higher SRS after an SL session preceded by a FAP session was 8/27 � .30, and the odds 8/19 � .42. The result is a significant odds ratio of 1.35/.33 � 4.93 (95% CI[1.71, 14.22]), FET p � .003, favoring FAP. The probability of a higher SRS after an SL session preceded by an SL session was 7/14 � .50, and the odds 7/7 � 1.0. The probability of a higher SRS after a FAP session preceded by a FAP session was 3/9 � .33, and the odds 3/6 � .50. The result is a nonsignificant odds ratio of 1.0/.50 � 2.00 (95% CI [0.35, 11.36]), FET p � .67. Postses- sion rating increases were most likely to occur when a FAP session followed an SL session.

On the FSBF, the NAP effect sizes were strong for 2 participants, moderate for 8 partic-

59ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

ipants, and weak for 3 participants. The weighted NAP analysis suggested a significant advantage for FAP, Z � 6.18, p � .001, with a moderate effect size, NAP � .75. The mean FSBF score was higher for the FAP sessions than SL sessions for 9/13 (69%). The .69 ob- served proportion favored FAP, but the bino- mial sign test failed to reach statistical signifi- cance, p � .27. Despite high means in both treatments, the FSBF data revealed a statisti- cally significant, small effect size, favoring FAP sessions (M � 36.47, SD � 5.62) over SL (M � 34.55, SD � 4.83), t(12) � �3.27, p � .01, g � .34. As with the WAI-SF and SRS, the FSBF data taken as a whole suggest a robust small to moderate effect favoring FAP sessions.

The probability of a higher FSBF after a FAP session preceded by an SL session was 26/40 � .65, and the odds 26/14 � 1.86. The probability of a higher FSBF after an SL session preceded by a FAP session was 9/28 � .32, and the odds 9/19 � .47. The result is a significant odds ratio of 1.86/.47 � 3.92 (95% CI [1.41, 10.93]), FET p � .01, favoring FAP. The probability of a higher FSBF after an SL session preceded by an SL session was 8/15 � .53, and the odds 8/7 � 1.14. The probability of a higher FSBF after a FAP session preceded by a FAP session was 5/9 � .56, and the odds 5/4 � 1.25. The result is a nonsignificant odds ratio of 1.14/1.25 � 0.91 (95% CI [0.17, 4.81]), FET p � 1.0. As was seen on the WAI-SF and SRS, increases on the FSBF were most likely to occur when a FAP session followed an SL session.

Hypothesis 3: Participants Would Report Improved Daily Life Interpersonal Relating With Changes Concentrated in Weeks Following FAP Sessions Compared With SL Sessions

The effects of FAP’s in-session relational focus are expected to generalize to client’s daily life interpersonal relationships. Paired samples t tests on the (pretreatment to posttreatment) FIS and MSIS were both significant, FIS, t(12) � 2.82, p � .02, and MSIS, t(12) � �2.43, p � .03, suggesting changes in out-of-session inter- personal relating. However, given the alternat- ing treatments design, these FIS and MSIS changes cannot be attributed exclusively to FAP.

The ORS was also collected prior to each session examining interpersonal functioning in the prior week. Paired samples t tests found the mean ORS score in the weeks following both FAP (M � 84.92, SD � 12.11) and SL (M � 85.15, SD � 12.66) sessions were statistically significantly higher than the ORS score from the week preceding the first therapy session (M � 71.40, SD � 14.91), FAP, t(12) � �3.28, p � .007, g � .93, and SL, t(12) � �4.12, p � .001, g � .93. However, when the mean ORS scores for weeks following FAP were compared with scores from weeks following SL, there was no difference, t(12) � 0.16, p � .88, g � .02. Thus, although participants reported improved inter- personal relating, it was not significantly better in the weeks following FAP.

Discussion

FAP is an interpersonal therapy wherein ther- apist-delivered in-session contingent conse- quences are the proposed mechanism of action. If this mechanism is differentially engaged in FAP compared with other treatments, FAP should produce stronger session ratings, owing to its distinctive focus. Using a restricted alter- nating treatments design and participants re- porting difficulties in interpersonal relating, the current study compared FAP with SL sessions. FAP sessions were marked by a greater imple- mentation of FAP rules designed to focus ther- apy on in-session behavior. In addition, postses- sion evaluations suggested robust small to moderate positive effects favoring FAP ses- sions, and that the sessions most likely to pro- duce an improved rating was a FAP session following an SL session. Thus, FAP was distin- guished from supportive therapy both from the perspective of an adherence coder and the per- spective of the participant receiving treatment. Both session types had substantial client- centered elements and were favorably evalu- ated. However, even in the face of these poten- tial ceiling effects, the average alliance and session ratings were superior following FAP sessions. These data are consistent with the FAP-generated hypotheses that in-session con- tingent responding contributes to stronger ther- apeutic alliances and meaningful sessions (Tsai et al., 2009), and provide critical support for the suggestion that the therapeutic alliance may serve as proxy for FAP’s mechanism of action

60 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

for use in efficacy research (Maitland & Gaynor, 2012, see also Gifford et al., 2011).

SL is an important comparison condition for FAP, as SL was designed to capture the stan- dard features of client-centered therapy thought to be responsible for building strong alliances. Nonetheless, alliance differences were found on measures that come from the common factors tradition and were designed to capture aspects of the alliance across all types of therapeutic approaches (Horvath, 2011; Rosenzweig, 1936). Theoretically FAP differentiates itself from client-centered approaches by specifying how the alliance contributes to the mechanism of action— by allowing the therapist to serve as a stimulus to evoke, shape, and reinforce behav- ior in the session. Thus, the proposed mecha- nism in FAP is being captured only by impli- cation in the differential alliance that is predicted from FAP. Microcoding of sessions using the Functional Analytic Psychotherapy Rating Scale (Callaghan, Follette, Ruckstuhl, & Linnerooth, 2008) coding system allows for more direct measurement of FAP’s proposed mechanism and has provided some promising data in small-N, single case studies (e.g., Busch, Callaghan, Kanter, Baruch, & Weeks, 2010). Microcoding is highly time and training inten- sive and its reliability and validity not fully established. As such, having a “middle level” unit such as the therapeutic alliance that could serve as a proxy appears useful for promoting larger scale efficacy research (Maitland & Gaynor, 2012), so long as it remains tied to basic behavioral principles on the laboratory science side, and clinical outcomes on the ap- plied science side (Vilardaga et al., 2009). Sup- portive efficacy data might be used to justify follow-up microprocess coding to more pre- cisely explore mechanism of action.

With respect to outcome, participants re- ported improved interpersonal relating on the FIS and MSIS. The current design did not con- trol for effects due to the passage of time, re- peated measurement, or simply being in any treatment condition. Other studies suggest FAP produces changes in interpersonal functioning that go beyond those attributable to relationship building alone (Landes, Kanter, Weeks, & Busch, 2013). In the current study, the ORS showed significant change over the course of treatment; however, differential change in the weeks following FAP sessions was not ob-

served. The lack of differential ORS change may be due to persisting effects of prior ses- sions that would result in ORS scores represent- ing a cumulative index (to that point), making it difficult to isolate therapy-specific effects on week-by-week basis. Another reason the ORS may not have changed differentially has to do with the assignment of homework. Ideal FAP homework follows directly from (and thereby promotes generalization of) improvements ob- served in session. Ideographic homework as- signments, based on hypothesized functional parallels between in-session contingencies and those occurring in the participant’s daily life, were not always readily available. As such, in 49% of FAP sessions, homework was not as- signed, meaning that explicit attempts to foster generalization during the next week could not be (or were not) offered. Thus, any generaliza- tion of behavior to daily life interactions would have to occur by less formal means, which would be expected to occur in FAP (Kohlenberg & Tsai, 1991), but may have been less obvious to the participant week-to-week than behaviors explicitly assigned for implementation.

The current findings set the stage for future efficacy research on FAP (see also Maitland & Gaynor, 2012). FAP was distinctive from SL, which methodologically supports a randomized clinical trial with FAP versus a supportive con- dition as the independent variable. FAP sessions also produced modestly stronger alliances than SL. Consequently, if FAP were compared with SL in a clinical trial, it is predicted that FAP would produce stronger alliances. Interpersonal relating improved over the course of treatment with FAP and SL, but these improvements could not be attributed to treatment or either session type, given the design. A randomized controlled trial would allow for effects to be more readily attributed to treatment and could test the prediction that FAP should outperform SL on social relating. It logically follows that the stronger alliance (driven by provision of FAP) should statistically mediate changes in daily life interpersonal relating (that follow from consequences applied to CRB in session). Such a finding would further support the sug- gestion that the differential therapeutic alliance is serving as a proxy for engagement of FAP’s mechanism of action. Of course direct measures of CRB would be preferable; however, prag- matically, in the absence of readily available

61ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

and psychometrically established indices, mea- sures of the therapeutic alliance may be valu- able in moving an efficacy research agenda for- ward (see Maitland & Gaynor, 2012).

There are a number of limitations. First, al- though the protocol offered 10 sessions, the mean number attended was eight. Fortunately, the mean was four sessions for both session types. Given our collegian sample, some opted to terminate participation when it coincided with the end of a semester and the participant’s permanent residence was a distance from cam- pus that made further participation untenable. It is also important to acknowledge there was only one therapist. This fact, along with the small, collegian sample, potentially limits generaliz- ability. Other practitioners may fair better or worse depending on their knowledge and skill applying SL and FAP, and the effects may differ with other populations. The therapist had participated in extensive training in FAP and was conceptually and clinically enamored with the approach. The role of allegiance effects are well known (and debated) in outcome research (Munder, Brütsch, Leonhart, Gerger, & Barth, 2013) and may have played a role in the current findings. However, the high alliance ratings in the SL sessions suggest that at least a minimum standard of client-centered engagement was ob- tained.

In conclusion, FAP sessions were distinctive from SL sessions in (a) the application of FAP rules focusing the therapeutic interaction on in- session contingencies, and (b) producing small to moderate improvements in the alliance and postsession evaluations. These improvements were most likely to occur when a FAP session followed an SL session. Thus, the results are consistent with a core supposition in FAP that the strong here-and-now focus should contrib- ute to particularly engaging therapy sessions and therapeutic relationships that are meaning- ful, intimate, and, possibly, curative (Tsai et al., 2009).

References

American Psychological Association. (2012). Reso- lution on the recognition of psychotherapy effec- tiveness. Retrieved from http://www.apa.org/about/ policy/resolution-psychotherapy.aspx

Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance.

Psychotherapy: Theory, Research & Practice, 16, 252–260. http://dx.doi.org/10.1037/h0085885

Busch, A. M., Callaghan, G. C., Kanter, J. W., Ba- ruch, D. E., & Weeks, C. E. (2010). The functional analytic psychotherapy rating scale: A replication and extension. Journal of Contemporary Psycho- therapy, 40, 11–19. http://dx.doi.org/10.1007/ s10879-009-9122-8

Busseri, M. A., & Tyler, J. D. (2003). Interchange- ability of the Working Alliance Inventory and Working Alliance Inventory, short form. Psycho- logical Assessment, 15, 193–197. http://dx.doi.org/ 10.1037/1040-3590.15.2.193

Callaghan, G. M., Follette, W. C., Ruckstuhl, L. E., & Linnerooth, P. J. N. (2008). The Functional Analytic Psychotherapy Rating Scale (FAPRS): A behavioral psychotherapy coding system. The Be- havior Analyst Today, 9, 98 –116. http://dx.doi .org/10.1037/h0100648

Clore, J. L., & Gaynor, S. T. (2012). Cognitive mod- ification versus therapeutic support for internaliz- ing distress and positive thinking: A randomized technique evaluation trial. Cognitive Therapy and Research, 36, 58 –71. http://dx.doi.org/10.1007/ s10608-010-9344-9

Corsini, R. J., & Wedding, D. (2005). Current psy- chotherapies (7th ed.). Toronto, Canada: Brooks/ Cole.

Descutner, C. J., & Thelen, M. H. (1991). Develop- ment and validation of a fear-of-intimacy scale. Psychological Assessment: A Journal of Consult- ing and Clinical Psychology, 3, 218 –225. http:// dx.doi.org/10.1037/1040-3590.3.2.218

Duncan, B. L., Miller, S. D., Sparks, J. A., Claud, D. A., Reynolds, L. R., Brown, J., & Johnson, L. D. (2003). The session rating scale: Preliminary psychometric properties of a “working” alliance measure. Journal of Brief Therapy, 3, 3–12.

Follette, W. C., & Bonow, J. T. (2009). The chal- lenge of understanding process in clinical behavior analysis: The case of functional analytic psycho- therapy. The Behavior Analyst, 32, 135–148.

Gaynor, S. T., & Lawrence, P. S. (2002). Comple- menting CBT for depressed adolescents with Learning through In Vivo Experience (LIVE): Conceptual analysis, treatment description, and feasibility study. Behavioural and Cognitive Psy- chotherapy, 30, 79 –101. http://dx.doi.org/10.1017/ S135246580200108X

Gifford, E. V., Kohlenberg, B. S., Hayes, S. C., Pierson, H. M., Piasecki, M. P., Antonuccio, D. O., & Palm, K. M. (2011). Does acceptance and rela- tionship focused behavior therapy contribute to bupropion outcomes? A randomized controlled trial of functional analytic psychotherapy and ac- ceptance and commitment therapy for smoking cessation. Behavior Therapy, 42, 700 –715. http:// dx.doi.org/10.1016/j.beth.2011.03.002

62 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Greenberg, L. S., Watson, J. C., & Lietaer, G. (Eds.). (1998). Handbook of experiential psychotherapy. New York, NY: Guilford Press.

Greenberg, R. P. (2012). Essential ingredients for successful psychotherapy: Effect of common fac- tors. In M. J. Dewan, B. N. Steenbarger, & R. P. Greenberg, (Eds), Art and science of brief psycho- therapies: An illustrated guide (2nd ed.). Arling- ton, VA: American Psychiatric Publishing.

Hanson, W. E., Curry, K. T., & Bandalos, D. L. (2002). Reliability generalization of working alli- ance inventory scale scores. Educational and Psy- chological Measurement, 62, 659 – 673. http://dx .doi.org/10.1177/0013164402062004008

Heyvaert, M., & Onghena, P. (2014). Randomization tests for single-case experiments: State of the art, state of the science, and state of the application. Journal of Contextual Behavioral Science, 3, 51– 64. http://dx.doi.org/10.1016/j.jcbs.2013.10.002

Horvath, A. O. (2011). Alliance in Common Factor Land: A view through the research lens. Research in Psychotherapy: Psychopathology, Process and Outcome, 14, 121–135.

Horvath, A. O., & Greenberg, L. S. (1989). Devel- opment and validation of the Working Alliance Inventory. Journal of Counseling Psychology, 36, 223–233. http://dx.doi.org/10.1037/0022-0167.36 .2.223

Ilardi, S. S., & Craighead, W. E. (1994). The role of nonspecific factors in cognitive-behavior therapy for depression. Clinical Psychology: Science and Practice, 1, 138 –155. http://dx.doi.org/10.1111/j .1468-2850.1994.tb00016.x

Kanter, J. W., Tsai, M., & Kohlenberg, R. J. (Eds.). (2010). Introduction to The Practice of Functional Analytic Psychotherapy. The practice of functional analytic psychotherapy (pp. 1–7). New York, NY: Springer. http://dx.doi.org/10.1007/978-1-4419- 5830-3_1

Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3, 1–27. http://dx.doi.org/ 10.1146/annurev.clinpsy.3.022806.091432

Kohlenberg, R. J. (2005). Functional analytic psy- chotherapy levels of integration. Retrieved from http://www.functionalanalyticpsychotherapy.com/ FAP%20Levels%20of%20Integration.pdf

Kohlenberg, R. J., & Tsai, M. (1991). Functional analytic psychotherapy: Creating intense and cu- rative therapeutic relationships. New York, NY: Plenum Press. http://dx.doi.org/10.1007/978-0- 387-70855-3

Kohlenberg, R. J., & Tsai, M. (1994). Functional analytic psychotherapy: A radical behavioral ap- proach to treatment and integration. Journal of Psychotherapy Integration, 4, 175–201. http://dx .doi.org/10.1037/h0101264

Lambert, M. J., Hansen, N. B., Umphress, V., Lun- nen, K., Okiishi, J., Burlingame, G., . . . Reisinger, C. W. (1996). Administration and Scoring Manual for the Outcome Questionnaire (OQ 45.2). Wil- mington, DE: American Professional Credential- ing Services.

Landes, S. J., Kanter, J. W., Weeks, C. E., & Busch, A. M. (2013). The impact of the active components of functional analytic psychotherapy on idio- graphic target behaviors. Journal of Contextual Behavioral Science, 2, 49 –57. http://dx.doi.org/10 .1016/j.jcbs.2013.03.004

Maitland, D. W. M., & Gaynor, S. T. (2012). Pro- moting efficacy research on functional analytic psychotherapy. International Journal of Behav- ioral Consultation and Therapy, 7, 63–71. http:// dx.doi.org/10.1037/h0100939

Martin, D. J., Garske, J. P., & Davis, M. K. (2000). Relation of the therapeutic alliance with outcome and other variables: A meta-analytic review. Jour- nal of Consulting and Clinical Psychology, 68, 438 – 450. http://dx.doi.org/10.1037/0022-006X.68 .3.438

Miller, R. S., & Lefcourt, H. M. (1982). The assess- ment of social intimacy. Journal of Personality Assessment, 46, 514 –518. http://dx.doi.org/10 .1207/s15327752jpa4605_12

Miller, S. D., Duncan, B. L., Brown, J., Sparks, J. A., & Claud, D. A. (2003). The Outcome Rating Scale: A preliminary study of the reliability, valid- ity, and feasibility of a brief visual analog measure. Journal of Brief Therapy, 2, 91–100.

Munder, T., Brütsch, O., Leonhart, R., Gerger, H., & Barth, J. (2013). Researcher allegiance in psycho- therapy outcome research: An overview of re- views. Clinical Psychology Review, 33, 501–511. http://dx.doi.org/10.1016/j.cpr.2013.02.002

Orlinsky, D. E., Rønnestad, M. H., & Willutzki, U. (2003). Fifty years of process-outcome research: Continuity and change. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (5th ed., pp. 307–390). New York, NY: Wiley.

Parker, R. I., & Vannest, K. (2009). An improved effect size for single-case research: Nonoverlap of all pairs. Behavior Therapy, 40, 357–367. http:// dx.doi.org/10.1016/j.beth.2008.10.006

Psychological & Educational Films. (Producer). (1981). Three approaches to psychotherapy I. pt. 1 [Therapist, Carl Rogers]. Corona Del Mar, CA: Author.

Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Jour- nal of Consulting Psychology, 21, 95–103. http:// dx.doi.org/10.1037/h0045357

Rosenzweig, S. (1936). Some implicit common factors in diverse methods of psychotherapy. American Journal of Orthopsychiatry, 6, 412–

63ALTERNATING FAP WITH SUPPORTIVE LISTENING SESSIONS

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

415. http://dx.doi.org/10.1111/j.1939-0025.1936 .tb05248.x

Rounsaville, B. J., Carroll, K. M., & Onken, L. S. (2001). A stage model of behavioral therapies re- search: Getting started and moving on from stage I. Clinical Psychology: Science and Practice, 8, 133–142. http://dx.doi.org/10.1093/clipsy.8.2.133

Siev, J., Huppert, J. D., & Chambless, D. L. (2009). The Dodo Bird, treatment technique, and dissem- inating empirically supported treatments. The Be- havior Therapist, 32, 69 –76.

Tsai, M., Kohlenberg, R. J., Kanter, J. W., Kohlenberg, B., Follette, W. C., & Callaghan, G. M. (Eds.). (2009). A guide to functional analytic psychotherapy: Awareness, courage, love, and behaviorism. New York, NY: Springer Science�Business Media. http:// dx.doi.org/10.1007/978-0-387-09787-9

Vannest, K. J., Parker, R. I., & Gonen, O. (2011). Single case research: Web based calculators for

SCR analysis (Version 1.0) [Web-based applica- tion]. College Station, TX: Texas A&M Univer- sity. Retrieved from http://www.singlecase research.org

Vilardaga, R., Hayes, S. C., Levin, M. E., & Muto, T. (2009). Creating a strategy for progress: A contex- tual behavioral science approach. The Behavior Analyst, 32, 105–133.

Weeks, C. E., Kanter, J. W., Bonow, J. T., Landes, S. J., & Busch, A. M. (2012). Translating the theoretical into practical: A logical framework of functional analytic psychotherapy interactions for research, training, and clinical purposes. Behavior Modification, 36, 87–119. http://dx.doi.org/10 .1177/0145445511422830

Received January 24, 2016 Accepted March 15, 2016 �

64 MAITLAND AND GAYNOR

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.