Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Journal of Contextual Behavioral Science

journal homepage: www.elsevier.com/locate/jcbs

Original Article

Whatever gets your heart pumping: the impact of implicitly selected

reinforcer-focused statements on exercise intensity
Marianne L. Jackson a,b,n, W. Larry Williams a, Steven C. Hayes a, Tiffany Humphreys a,
Brook Gauthier a, Ryan Westwood a
a
Department of Psychology, University of Nevada, 1664 N Virginia St., Reno, NV 89557, USA
b
Department of Psychology, California State University, 2576 E. San Ramon Ave. M/S/ ST11, Fresno, CA 93740-8039, USA

art ic l e i nf o a b s t r a c t

Article history: This study investigated the impact of reinforcer-focused statements vetted by the Implicit Relational
Received 4 February 2013 Assessment Procedure (IRAP) on the intensity of participation in an exercise class. Using a relative heart
Received in revised form rate measure, the first experiment compared individually selected phrases focused on the positive
24 August 2015
consequences of exercise to general topographic advice about how to exercise using an alternating
Accepted 5 November 2015
treatments design, and found that participants exercised harder in the presence of the reinforcer-focused
phrases. Experiment 2 tested a similar comparison using phrases focused on consequences that were
Keywords: shown to be neutral or negative by the IRAP, and found no consistent effect on responding. These results
Motivative augmentals support a key prediction from previous research in Relational Frame Theory, namely that reinforcer-
Motivation
focused statements can function as motivative augmentals, and they suggest that such motivative effects
Transformation of stimulus functions
can impact behaviours of applied importance such as exercise.
Relational Frame Theory
Exercise & 2015 Published by Elsevier Inc. on behalf of Association for Contextual Behavioral Science.
Fitness

1. Introduction A to stimulus B and from stimulus B to stimulus C are learned, and

the relations from stimulus B to stimulus A and stimulus C to
It is important to know what motivates people, particularly stimulus B are entailed, then the relation from stimulus A to sti-
during difficult behavioural tasks such as studying, exercising, or mulus C and from stimulus C to stimulus A can be derived), and
maintaining a diet. Consequences can strengthen such actions, but transformation of stimulus function (the functions of a stimulus in
as a technical matter, motivational events happen before con- a relational frame may be transformed in accordance with the
sequences are experienced, altering the effectiveness of con- underlying relation between that stimulus and the other stimuli in
sequences (what is known in technical terms as an “establishing the frame). For example, if a person learns that Oscar the dog is
operation” (Michael, 1982, 1993)). Verbal statements such as goals, bigger than Felix but smaller than Max, that person can also derive
values, or descriptions of consequences often appear to function as that Felix is smaller than Oscar and that Max is bigger than Oscar
motivational events, but relatively few studies have focused on the (mutual entailment), and that Felix is smaller than Max and Max is
content of such statements as they relate to consequent events. bigger than Felix (combinatorial entailment). If Oscar bites the
Relational Frame Theory (RFT) has provided a way to think of person, resulting in a fear response, the person may now be more
such motivational effects. The central idea of RFT is that human afraid of Max and less afraid of Felix, due to the underlying relation
language and cognition is dependent on learned relational re- of comparative size (transformation of function). Many different
sponses that can be regulated by arbitrary contextual cues (Hayes, stimulus functions have been shown to be transformed via derived
Barnes-Holmes, & Roche, 2001). These are referred to as relational relational responding including discriminative control (Roche,
frames and are defined by the properties of mutual entailment (if a Barnes-Holmes, Smeets, Barnes-Holmes, & McGeady, 2000), con-
unidirectional relation from stimulus A to stimulus B is learned ditioned reinforcement and punishment (Greenway, Dougher, &
then a second unidirectional relation from B to A is entailed), Wulfert, 1996), respondent elicitation (Dougher, Auguston, Mark-
combinatorial entailment (if unidirectional relations from stimulus ham, Greenway, & Wulfert, 1994) mood (Barnes-Holmes, Barnes-
Holmes, Smeets, & Luciano, 2004), and many others. Transforma-
n tion of stimulus functions is central to the RFT concept of moti-
Corresponding author at: Department of Psychology, California State University,
2576 E. San Ramon Ave. M/S/ ST11, Fresno, CA 93740-8039, USA. vative augmentals that “temporarily alter the degree to which
E-mail address: [email protected] (M.L. Jackson). previously established consequences function as reinforcers or

http://dx.doi.org/10.1016/j.jcbs.2015.11.002
2212-1447/& 2015 Published by Elsevier Inc. on behalf of Association for Contextual Behavioral Science.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
2 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

punishers” (Hayes et al., 2001, p.109). alternative relational terms and asks participants to respond
Transformation of stimulus functions can alter motivation in at quickly and accurately given feedback that is either consistent or
least two ways. One way is to describe, or verbally invoke, states of inconsistent with a possible relation; latency differences between
deprivation or aversive stimulation that would function as moti- consistent and inconsistent trials indicate the strength of parti-
vators, relying on the transformation of stimulus functions to es- cular verbal relations. For example, “ice cream” and “yummy”
tablish verbal equivalents of these direct motivational conditions. might be presented as stimuli and participants might be asked to
For example, Valdivia, Luciano, and Molina (2006) showed that select either same or opposite from the row below. People with
children became thirsty and drank more water when presented
well-practiced preferences for ice cream will generally have
with stories about being very hot in a desert; similarly, a story
shorter latencies and fewer errors selecting same than opposite.
about being physically restricted in a box led to more stretching.
Research has shown that the IRAP is difficult to fake (McKenna,
A second way to increase motivation is to present the percep-
Barnes-Holmes, Barnes-Holmes, & Stewart, 2007). It has been
tual or sensory functions of a reinforcer, which can serve to mo-
shown to detect a wide variety of implicit attitudes in such areas
tivate behaviour that produces it, similar to physically providing a
as race, religion and gender (Drake et al., 2010), autism (Milne,
small sample of a reinforcing consequence – a motivational pro-
cedure that is known to be powerful (Ayllon & Azrin, 1968; see also Barnes-Holmes, Barnes-Holmes, & Stewart, 2005), and age (Cullen,
Skinner (1936) for an example of the same effect created via Barnes-Holmes, Barnes-Holmes, & Stewart, 2009). Implicit atti-
classical conditioning). Ju and Hayes (2008) showed in a laboratory tudes in turn have been shown to predict behaviours of applied
experiment that children presented with stimuli in an equivalence importance, such as success or failure in drug treatment programs
class with an appetitive consequence showed higher rates of in- (Carpenter, Martinez, Vadhan, Barnes-Holmes, & Nunes, 2012).
strumental behaviour directed toward that consequence. The The present study examined the motivative effect of reinforcer-
motivative effects of such reinforcer-focused statements were ex- focused statements on the intensity of exercise. Exercise is uni-
amined in the present study. versally agreed to be an important health behaviour (Blair, 2009;
What RFT adds to a contextual behavioural interpretation of Naci & Ioannidis, 2013) but there is relatively little research on
motivation is that there are verbal parallels to known establishing how to impact its intensity. That is a significant omission in the
operations, such as deprivation or contact with reinforcers, which literature, since peak intensity predicts the cardiovascular impact
operate via a transformation of stimulus functions through rela- of exercise sessions much more than does the volume of training
tional networks. This is an idea with clear applied implications but (Winett, 1998).
to date there is no formal research on the applied impact of mo- There is a small body of research that has examined the role of
tivative augmentals as analysed within RFT. implicit attitudes on other aspects of exercise. Implicit measures
In a general way, advertisers seem to rely on these effects when have been shown to be better predictors of attitudes toward ex-
they verbally remind listeners of positive sensory qualities of their
ercise and health behaviours than explicit measures (Huang &
products (e.g., how cold beer is or sweet ice cream is), or of the
Hutchinson, 2008). They predict intentions to be physically active,
conditions of deprivation or aversive stimulation that might mo-
which in turn predicts changes in exercise behaviour (Hagger,
tivate their acquisition (e.g., how hot it is; how thirsty you are). A
Chatzisarantis, & Biddle, 2002). Additional research has shown
central task of the present study was to examine the motivational
impact of reinforcer-focused verbal statements. In applied settings, that both implicit and explicit motivation are relevant to exercise:
however, more individualized knowledge is needed about the implicit measures predict the preferred properties of exercise and
particular consequences that are most positive during specific intention to engage in physical activity, and explicit measures
behavioural tasks. These depend on details of an individual's his- predict the likelihood that these intentions will result in a change
tory that may not be accessible. in actual exercise behaviour and in the initiation of exercise
Direct self-report is the most common way to attempt to access (Hagger, 2012). Researchers in this area have called for greater
such histories, but there is a great deal of research to suggest that attention to implicit measures while pointing out that we do not
such measures are susceptible to social influences that vary de- yet know much about how procedures based on these measures
pending on the context in which self-report occurs. In light of this, might have their effects (Berry, Jones, McLeod, & Spence, 2011).
measures such as the Implicit Associations Test (IAT; Greenwald, The consequences of physical exercise can often be delayed,
McGhee, & Schwartz, 1998) have been developed to assess eva- highly social, and idiosyncratic. Some people may be exercising to
luative responses that appear to occur rapidly and automatically. lose weight; others may be doing so for their cardiovascular health,
These responses are commonly referred to as implicit responses, still others to live a long life. RFT suggests that statements that link
whereas the former are referred to as explicit responses. Implicit present exercise to these consequences could provide a motiva-
and explicit measures converge or diverge, depending on the sti- tional effect, but only if they fit what the person strongly desires.
muli involved.
The present study examined this idea by assessing the impact of
One limitation of methods such as the IAT is that they focus on
implicit statements related to consequences of individual im-
the assessment of associations between stimuli and not the di-
portance, on exercise intensity. If statements of this kind were
rectionality or nature of the relations between them. This may not
found to be useful, a next step would be to compare them to explicit
be a significant limitation for approaches that consider associa-
self-reports of relevant consequences for their relative utility.
tions to be at the core of language and cognition, but it is a sig-
In the context of a university-based fitness class, verbal stimuli
nificant issue for RFT as it considers all forms of language and
cognition to be relational acts. were presented that were implicitly related to either preferred or
RFT researchers have accordingly developed an implicit mea- non-preferred consequences of exercise for individual participants
sure that better comports with RFT conceptions: the Implicit Re- as identified by the IRAP. The effects of these statements were
lational Assessment Procedure (IRAP; Barnes-Holmes et al., 2006). compared over two studies to the effects of verbal exhortations
The IRAP allows for the specific assessment of existing relational about the topography of the exercise needed in the class. Explicit
frames, and specification of the nature and direction of the rela- self-reports of preference were also collected, but no experimental
tions involved. The IRAP presents two stimuli and a set of procedures were linked to them.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

2. Experiment 1 Table 1
A list of all phrases used for IRAP-1 in Experiments 1 and 2.
2.1. Method
Questionnaire Participant-provided

2.1.1. Participants and setting Exercise will reduce my risk of cancer. To get in better shape
Participants in this experiment were 11 female undergraduate Exercise will help me lose weight. To improve endurance
students enroled in an indoor cycling class at the University of Exercise will help manage my blood Better health
pressure.
Nevada, Reno. The experimenters recruited participants by reading Exercise will make me thinner. Prevent disease
a description of the procedures aloud to the class; however, it Exercise will boost my immune system. Feel better about myself
should be noted that the explicit hypothesis was not revealed to Exercise will tone my muscles. To feel more confident about
participants until completion of the study. The Institutional Re- myself
Exercise will improve my circulation. To be in better shape
view Board (IRB) approved the study and all participants com-
Exercise will help me look better. To fit into my nice jeans
pleted informed consent. Then participants completed all paper- Exercise will reduce my risk of heart To feel better about body
and-pencil questionnaires in the exercise room. Next participants disease.
were given both computerized IRAP assessments in a computer lab Exercise will improve my respiratory To lose fat
system.
in the university library. The first of these will be referred to as
Exercise will make me feel more Improve physical wellbeing
IRAP-1 and the second as IRAP-2. Finally, each participant com- attractive.
pleted 10 class sessions that varied in duration from 31 to 70 min Exercise will make my clothes fit better. Weight control
(mean¼ 45 min), with no more than one session conducted per To wear my skinny jeans
day. All cycling classes were held in the recreation building in a
room containing approximately 30 stationary bikes and were
very important to not important for them, and were also asked to
taught by a certified group fitness instructor employed by the
add any specific reasons of their own that had not been included
university.
but that they felt were important. An example of a participant-
provided reason was “to fit in my skinny jeans.” A total of six
2.1.2. Design
phrases from each category (health and aesthetic) were used to
Experiment 1 used an alternating treatments design (ATD;
Barlow & Hayes, 1979) that called for session-by-session alterna- create the IRAP-1 for each participant, including any idiosyncratic
tion of a motivational phrase condition with an instructional reasons they had added and the most highly ranked reasons from
phrase condition. All participants began with three baseline ses- those provided.
sions followed by three alternating sessions, one return to base-
line, and three more alternating sessions. The order of treatments 2.1.6. IRAP
was counterbalanced across participants: during any given class in Participants completed two computerized versions of the IRAP
the alternating conditions, half of the participants were in the (2006) (downloaded from http://www.irapresearch.org), which at
motivational condition, and half were in the instructional the time this research was conducted was the most current ver-
condition.. sion. Some details of the IRAP procedures depart from those used
No statements were given to participants during baseline in current studies but were consistent with IRAP research pub-
conditions. During the motivational condition, participants re- lished at the time of the study (Barnes-Holmes et al., 2006;
ceived statements referring to their own preference on the IRAP-1 Hughes & Barnes-Holmes, 2013).
except for those who did not show an effect on the IRAP-1. These The first IRAP (IRAP-1) utilized phrases from the Reasons to
participants were given an equal number of aesthetic and health Exercise questionnaire, including individual participant-provided
related statements randomly selected from IRAP-2 stimuli. The reasons, as the target phrases. A label was presented in the top-
phrases provided during the instructional condition were state- centre of the screen that said either, “I exercise” or “I don't ex-
ments derived from the instructor about the desired topography of ercise,” followed by a target phrase from the questionnaire, and
the exercises (e.g., “keep your shoulders down and push through two response options of “true” or “false.” A complete list of the
your heels”). phrases drawn from the Reasons to Exercise questionnaire and all
participant-provided reasons is shown in Table 1. Each in-
2.1.3. Assessments dividualized IRAP-1 included 12 phrases, 6 health-related phrases
All participants were given the following assessments, prior to and 6 aesthetic-related phrases. A sample of stimuli used for one
baseline data collection. The researchers created the Exercise participant can be seen in Table 2. Given the results of the forced-
History questionnaire and the Reasons for Exercise questionnaire choice question across participants, all IRAP-1s were programmed
for the purposes of this study. with the hypothesis that participants primarily exercised for more

2.1.4. Exercise history Table 2

Sample of labels, target phrases, and response options for a given participant’s
Participants were asked to complete a brief questionnaire
IRAP-1 in Experiments 1 and 2.
about their history of exercise and fitness, and their views on their
own current and previous levels of fitness attained. Participants Label 1: I Exercise Label 2: I don’t exercise
were then asked to indicate by checking a box whether health or Consistent target phrases 1: To Consistent target phrases 2: to boost my
lose weight immune system
aesthetic reasons were their overall purpose in exercising. This
to tone muscles to manage my blood pressure
provided a force-choice measure of their overall reason and was to look better to improve my respiratory system
used to guide the development of the programmed hypothesis for to be thinner to reduce risk of heart disease
the IRAP-1. to feel more attractive to improve my circulation
so that clothes fit better to reduce my risk of cancer
2.1.5. Reasons for exercising Response Option 1: True Response Option 2: False

Participants were asked to respond to a list of 12 health and

Note: On any one trial, only one label was presented with one target phrase, and
aesthetic reasons to exercise (see Table 1 for the phrases used). the participant was required to respond to one of the two response options
Participants rated each reason on a 4-point Likert-type scale from provided.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
4 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

immediate aesthetic reasons and not for more delayed health from the mean latency for inconsistent test blocks. Cutoffs for
reasons. This meant that responses to aesthetic reasons as true individual use of the IRAP suggested by Milne et al. (2005) were
were considered correct during consistent trials and incorrect applied: participants who responded more rapidly by an average
during inconsistent trials, and vice versa for health reasons. Par- of 40 ms or more to trials that were consistent with the pro-
ticipants were not informed of the programmed hypothesis and grammed hypothesis were categorized as having aesthetic reasons
the consistent and inconsistent trial blocks were presented in an (or, in the IRAP-2, both health and aesthetic reasons); if their
alternating fashion. Each assessment involved two practice blocks scores were  40 ms or less, they were categorized as having
followed by six test blocks, and each block consisted of 24 trials. If health reasons (or, in the IRAP-2 direct sensory reasons); those in
a participant failed to achieve at least 80% correct on either between were classified as not having shown an IRAP preference.
practice block they were required to repeat the practice blocks More current methods of analysis such as D-IRAP scores, which
(only two of the participants repeated the practice blocks and both control for individual differences in age, motor skills, and cognitive
passed with just one repetition). No participants dropped out due ability (Barnes-Holmes et al., 2010), were not deemed to be nee-
to a failure to respond correctly. ded for post-analysis because only ordinal IRAP results within
Responses that were consistent with the programmed hy- participants were used. No group level IRAP data were used and no
pothesis were followed by a 400 ms intertrial interval and the parametric differences across individuals were examined. Milne's
initiation of the next trial. Incorrect responses were followed by approach was used because it is a single-subject approach that fit
computerized feedback consisting of a large red X on the screen the purposes of the study, is easy to score, and fits the practical
and the opportunity to correct the response by selecting the al- realities of health and fitness professionals.
ternative response option. Once the participant had made the
correct response, a 400 ms intertrial interval occurred followed by 2.2.2. Exercise class
the start of the next trial. At the end of each trial block, partici- The instructor and all participants wore a Polar RS400s heart
pants were instructed to reverse their answers. rate monitor during classes. Readings of average heart rates were
The second IRAP (IRAP-2) compared health and aesthetic rea- taken after class from each monitor. Because instructors cycled
sons with the direct consequences of exercise and was used pri- with the class members and deliberately varied the intensity of
exercise session to session, the dependent measure was the par-
marily for control purposes. Participants were asked to respond to
ticipant's average heart rate for that session divided by the in-
them as “good” or “bad.” This IRAP was programmed with the
structor's average heart rate for the same session. This allowed the
hypothesis that delayed health and aesthetic reasons such as “burn
intensity of each class to vary in a manner determined by the in-
fat” or “lose weight” were more positive consequences of exercise
structor, while accounting for such variability by calculating the
than immediate and direct sensory consequences such as
extent to which participants met the instructor-set heart rate
“breathing heavily.” This IRAP was not individualized and thus
goals.
presented the same labels, target phrases, and response options
for all participants. These can be seen in Table 3. Each IRAP took
2.2.2.1. Procedure. Students were told by the class instructor to
approximately 15–20 min to complete and both were completed
keep their heart rate within a specified range that varied according
during one session.
to the intensity target of a particular class session (this varied and
was not controlled by the experimenter). Heart rate goals (e.g., 70–
2.2. Dependent variables 75% of maximum) were the same for all students at any given time
and the instructor stayed at the upper range of the heart rate range
2.2.1. IRAP provided to the class. During motivational and instructional pha-
The dependent variable in the IRAP was response latency, ses, all participants were given a card before class containing a
measured in ms. All trial latencies longer than 3000 ms were re- brief written statement that they were asked to remember and
coded as 3000 ms, and all latencies shorter than 300 ms were think about during class. The specific statements varied among
recoded as 300 ms. More recent IRAP research has tended to uti- individuals and the instructor did not know the statements used
lize a 2000 ms criterion since it appears to increase the validity or their assignment to individuals. The instructor did not receive a
and reliability of IRAP performance (Barnes-Holmes, Murphy, card in any sessions but at 5 min intervals reminded all partici-
Barnes-Holmes, & Stewart, 2010) but for longer phrases a 3000 ms pants to think about the words on their card for that day.
criterion is still commonly used, in line with the suggestion of
Barnes-Holmes, Barnes-Holmes, Stewart, and Boles (2010) to ad- 2.2.2.2. Interobserver Agreement (IOA) and Procedural Reliability
just the criterion to suit the population and complexity of the (PR). During 50% of all sessions a second research assistant con-
stimuli. ducted reliability checks on data collection from the heart rate
The mean latency for consistent test blocks was subtracted monitors. Cards were also checked to make sure they contained
the correct statements and reminders from the instructor to check
Table 3 the cards were recorded. IOA and PR were calculated using point-
Labels, target phrases, and response options for all participants’ IRAP-2 in Experi- to-point correspondence: the number of agreements was divided
ments 1 and 2.
by the number of agreements plus disagreements and multiplied
Label 1: Exercise GOOD Label 2: Exercise BAD by 100. IOA on the recording of heart rate measures from the
Consistent target phrases 1: lose Consistent target phrases 2: muscles monitors was 100%. Procedural integrity for correct cards was
weight hurt 100%, and procedural integrity for the instructor's reminders to
tone body get sweaty
think about the phrases was 98%.
be healthier hard to breathe
burn fat face gets red
prevent disease gets hot 2.3. Results
improve heart health legs get tired
Response Option 1: Same Response Option 2: Different 2.3.1. Assessments
Note: On any one trial, only one label was presented with one target phrase, and
2.3.1.1. Exercise history. None of the participants reported cur-
the participant was required to respond to one of the two response options rently being fit. On a scale of 1–4, with 1 being “currently fit” and
provided. 4 being “not fit”, participants reported a mean score of 2.9, with a

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Table 4 showed no differentiation.

Questionnaire and IRAP data from Experiment 1.
2.3.1.3. Exercise class. The average heart rates of each participant as
Pt. Standard Idiosyncratic Forced IRAP 1 IRAP 2 Reasons used
reasons reasons choice a percentage of the instructor's average heart rate during each
component are shown in Fig. 1, along with the overall average.
1 Health Both Health Health H/A Health Fig. 2 shows the average change from baseline for each participant
2 Both Both Aesthetic Aesthetic No effect Aesthetic in each condition. An ATD is typically analysed visually by the lack
3 Health Health Aesthetic Health No effect Health
4 Aesthetic Aesthetic Aesthetic Aesthetic H/A Aesthetic
of overlap between conditions — 9 of the 10 participants who
5 Aesthetic Health Aesthetic Aesthetic H/A Aesthetic showed an IRAP preference showed consistently higher relative
6 Both Aesthetic Aesthetic Aesthetic H/A Aesthetic heart rates in the motivational condition than in the instructional
7 Both Both Aesthetic Aesthetic H/A Aesthetic or baseline conditions. Participant 10 did not show an IRAP effect
8 Health Both Aesthetic Health H/A Health
and did not show any clear differentiation in heart rate measures
9 Both Health Aesthetic Aesthetic H/A Aesthetic
10 Health Aesthetic Aesthetic No effect No effect Both across conditions. Participant 8 showed a health preference on the
11 Health Health Health Health H/A Health IRAP, and self reported a preference for both types of reasons, but
also did not show a motivational effect.
“H/A” means that health and aesthetic reasons were preferred over direct sensory These data were analysed statistically using mixed effects
consequences in the IRAP 2.
models to examine the average outcomes for conditions, including
a random effect to account for the within subject correlation of all
measures within a subject, and a first-order autoregressive cov-
ariance parameter within condition (Gibbons, Hedeker, Elkin, &
range of 2–4. Nine of the participants reported that they had, at
Waternaux, 1993). This approach adjusts for the fact that measures
one time considered themselves to be physically fit, and two sta-
acquired nearer each other in time may be more correlated than
ted that they had never been fit. Participants reported various
points further away in time, and that this parameter of auto-cor-
current frequencies of exercise, with a mean of 2.9, and a range of
relation may vary by condition. Degrees of freedom were esti-
0–5 times per week. Any form of activity that participants con-
mated with the Kenward-Roger's (1997) approximation, which
sidered to be physical exercise was included as an occurrence. The
accommodates small sample inferences. All analyses were con-
results of the forced choice of overall purpose showed that nine of
ducted using SAS version 9.2.
the 11 participants reported that they exercised for aesthetic rea-
Results showed an overall difference between the three con-
sons. These results are shown in Table 4.
ditions (F(2, 20) ¼23.08, p o0.0001, d ¼2.15). Contrasts showed
that the motivational statements were significantly superior to
2.3.1.2. Reasons for exercising and IRAP. Categorization of partici-
both baseline (t(16) ¼  5.99, po 0.001, d ¼2.99) and instructional
pants based on self-report and the IRAP are shown in Table 4. The
conditions (t(26) ¼5.87, p o0.001, d ¼2.30). The contrast of base-
decision rules for the IRAP were linked to experimental proce-
line and instructional conditions was not statistically significant (t
dures and have already been described. In order to provide some
(19) ¼1.52, p ¼0.14, d ¼0.70).
overall sense of the self-report data, reasons for exercise are ca-
tegorized as health if there were a greater number of health rea-
2.4. Discussion
sons rated as important or very important than aesthetic reasons
rated as such. For example if a participant rated four out of the six
The results of Experiment 1 suggest that presentation of verbal
health reasons as important and only one of the aesthetic reasons
statements focused on individual-specific reinforcers for exercise
as such, their responses would be categorized as health. Similarly,
immediately increases objectively assessed exercise intensity. Re-
if six of the 10 idiosyncratic reasons given were health related,
inforcer-focused statements vetted by the IRAP were associated
responses to this section would be categorized as health. If the
with more intense exercise for 9 of the 11 participants in the study,
participant reported or rated an equal number of health and aes-
while 8 of these 9 participants also engaged in less intense ex-
thetic reasons as important or very important, their responses
ercise in the presence of the instructional statements. Evaluated at
were categorized as “both”.
the level of the entire group, reinforcer-focused statements were
Across all three self-report methods (rating standard measures,
significantly different overall than both baseline and instructional
idiosyncratic reasons, and forced choice on overall reason), the
statements, while the latter two conditions did not differ.
outcomes were consistent across measures for two participants,
It is possible, however, that the apparent motivational effects of
with participant 4 consistently reporting aesthetic reasons and
the reinforcer-focused statements occurred because they referred
participant 11 consistently reporting health reasons; the IRAP-1
to consequences of any kind, not to idiosyncratic reinforcers
confirmed these for both participants. Nine participants showed a
per se. Experiment 2 examined this possibility. All aspects of the
combination of health and aesthetic outcomes across self-report
experiment were kept constant, except that the phrases referred to
measures, seven of whom showed a dominant outcome (two
consequences that were shown to be non-preferred or less pre-
health and five aesthetic) and two of whom did not. The IRAP-1
ferred by the IRAP. If the results of Experiment 1 occurred because
confirmed the dominant outcome for six of these seven partici-
consequential phrases of any kind were preferred to instructional
pants, while one participant (participant 10) showed no differ-
statements, then neutral or negative exercise consequences as
entiation. The remaining two participants who did not have a
assessed by the IRAP should still motivate increased performance.
dominant outcome across self-report measures did show an IRAP
effect with participant 8 showing a preference for health reasons
and participant 9 showing a preference for aesthetic reasons.
3. Experiment 2
Overall the IRAP-1 confirmed the dominant reports of eight par-
ticipants, showed no effect for one, and clarified the preference for
3.1. Method
two who appeared indifferent in the self-report measures.
The IRAP-2 showed that eight of the participants classified
3.1.1. Participants and setting
health and aesthetic reasons as positive outcomes of exercise
Four female undergraduate students enrolled in an indoor cy-
when compared to other effects such as breathing hard; the others
cling class participated. This class lasted for an average of 50 min

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
6 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Overall Average Results

104 Motivation

Percent of Instructor’s Heart Rate

100
Baseline
96

92
Instruction
88

1 2 3 4 5 6 7 8 9 10
Class

Participant 1 Participant 2 Participant 3

105 Motivation 105 Motivation 105 Motivation

100 Baseline 100 Baseline 100

Baseline
95 95 95

90 90 90
Instruction
85 85 85
Instruction
80 80 80 Instruction

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Participant 4 Participant 5 Participant 6

105 Motivation 105 Motivation 105 Motivation
Baseline
100 Baseline 100 100
Percent of Instructor’s Heart Rate

95 95 95 Baseline
90 90 90
Instruction
Instruction
85 85 85
Instruction
80 80 80

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Participant 7 Participant 8 Participant 9

105 Motivation 115 Baseline 105
Baseline Motivation
Baseline
100 110 100

95 105 95
Instruction Motivation
90 100 90

85 Instruction 95 85
Instruction
80 90 80

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Participant 10
110 Participant 11
105 Instruction 105

100 100 Motivation

Baseline
95 Motivation 95
Baseline
90 90
Instruction
85 85

80 Implicitly Positive 80

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Class
Fig. 1. Data averaged across participants for each condition (top), and individual participant data, for Experiment 1.

(range of 42–56 min), and each participant completed 10 sessions 3.1.2. Design, procedures, independent, and dependent variables
during this experiment. All other features of the class and the The design, procedures, and all dependent variables were the
setting were the same as in Experiment 1. same as Experiment 1. The only difference was that the statements

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

3.2.1.3. Exercise class. The average heart rates of each participant as

a percentage of the instructor's average heart rate during each
component are shown in Fig. 3, along with the overall average.
Fig. 4 shows the average change from baseline for each participant
in each condition. There were no consistent differences between
conditions for any of the four participants. Participant 12 showed a
weak and inconsistent motivation effect for aesthetic reasons,
which was the type most clearly identified by self-report but not
by the IRAP-1. Participant 13 was also given statements related to
the non-preferred outcomes according to the IRAP-1 but showed
no differentiation in conditions. Participants 14 and 15 were pre-
sented with direct outcomes of exercise from the IRAP-2 during
the motivational condition — neither demonstrated a significant
difference in responding across conditions. Overall, these data
show that statements referring to neutral or negative con-
sequences, as assessed by the IRAP, did not motivate increase ex-
ercise intensity. This, in turn, strengthens the initial interpretation
of Experiment 1: implicitly selected reinforcer-focused statements
functioned as motivative augmentals, increasing exercise intensity.

4. General discussion
Fig. 2. Participants’ average change from baseline during each condition, for Ex-
periment 1. This is the first study to have shown that reinforcer-focused
verbal statements serve as motivational stimuli, altering objec-
used in the motivational condition referred to consequences that tively measured and difficult health behaviours. This supports a
were identified by the IRAP-1 (for two participants) or IRAP-2 (for key prediction of RFT (see, for example, Barnes-Holmes et al.,
two participants) as being less preferred or non-preferred. 2001): that such statements may function as motivational aug-
mentals due in part to the derived transformation of consequential
3.1.3. Interobserver Agreement (IOA) and Procedural Reliability (PR) stimulus functions. Such an effect has been shown in the labora-
During 62% of all sessions, reliability checks were conducted on tory (Ju & Hayes, 2008), but not in application. The results of this
recordings of heart rate measures from the monitors, cards given study suggest that similar effects are possible.
to participants, and reminders given by the instructor. IOA on The present studies used the IRAP as a means of selecting re-
heart rate measures was 98%. PR for cards given and instructor inforcer-focused motivational statements for each participant.
reminders was 100% and 96% respectively. Overall, the results suggest that the IRAP performed that task well.
Experiment 1 found that IRAP-vetted phrases had a motivational
3.2. Results effect as compared both to baseline and to instructions focused on
the proper topography of exercise. Experiment 2 eliminated the
3.2.1. Assessments possibility that this effect was due to the use of consequential
3.2.1.1. Exercise history. None of the participants reported cur- phrases of any kind or to a simple contrast effect as compared to
rently being fit. Three of the four participants reported that they instructional phrases.
had previously been fit, and all participants reported that they It could be argued that the motivational statements may have
currently exercised an average of three times per week. served a discriminative function, not a motivational function, and
this cannot be entirely ruled out.
3.2.1.2. Reasons for exercising. Categorization of participants based This would require that the motivational statements signaled
on self-report and the IRAP are shown in Table 5. Self reported the availability of the relevant consequences for exercise beha-
reasons for exercise varied depending on the specific self-report viour, but that the other statements, or the absence of statements,
measure. Participants 12 and 13 received their non-preferred did not (or did so to a lesser degree). In other words, either
reasons (health or aesthetic) according to the IRAP-1. For Partici- through verbal relations or a direct behavioural history, exercise in
pant 12 this corresponded with their checked overall self-reported the presence of the motivational statement, for example lose
preference, but this was not the case for Participant 13. Partici- weight, was somehow more likely to lead to weight loss, when
pants 14 and 15 received statements linked to the non-preferred compared to exercise behaviour in the presence of the instruc-
direct sensory consequences of exercise (e.g., breathing heavily) as tional or the non-preferred phrases (as in experiment 2). It seems
identified by the IRAP-2. unlikely that this discriminative function was established by the
experimental procedures, as participants had a similar experi-
Table 5 mental history with all types of statements used.
Questionnaire and IRAP data from Experiment 2. The IRAP and self-report measures suggest that the motiva-
tional phrases participated in some type of verbal relation, possi-
Pt. Standard Idiosyncratic Checked IRAP 1 IRAP 2 Reasons used
bly a causal relation, with exercise prior to the experiment. It is
reasons reasons choice
also possible that the motivational statements had a dis-
12 Both Health Aesthetic Health H/A Aesthetic criminative function, prior to the experiment. If this were the case,
13 Aesthetic Aesthetic Aesthetic Aesthetic H/A Health the difference between the motivational condition and other
14 Aesthetic Health Health Health H/A Direct conditions would suggest that the motivational statements might
15 Health Health Health Health H/A Direct
have enhanced existing discriminative, or evocative, functions. In
“H/A” means that health and aesthetic reasons were preferred over direct sensory either case, it seems that the functions of the words in the state-
consequences in the IRAP 2 ments are important.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
8 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 3. Data averaged across participants for each condition (top), and individual participant data, for Experiment 2.

Ju and Hayes (2008), described a similar motivational effect as preferences varied across questionnaire types for 11 of the 15
the consequential stimulus function hypothesis. This would suggest participants in both experiments, based on whether the focus was
that the words in the motivational statements had acquired some standard reasons for exercise, idiosyncratic reasons for exercise, or
of the reinforcing functions of the outcomes or events they are the checked overall preference (see Tables 4 and 5). There was
related to via a transformation of stimulus function. Presentation general correspondence between the IRAP-1 and checked overall
of these words provides a type of brief derived contact with those preference on the forced choice, however, which agreed for 11 of
reinforcing functions, increasing the value of those outcomes and the 15 participants. The differences are too few to determine
evoking (exercise) behaviour causally related to them. This would whether the IRAP did a better or worse job of selecting motivating
suggest that the motivational statements served a motivational statements than would self-report, especially because the four key
function, enhancing the momentary effectiveness of the related participants showed somewhat inconsistent results. Participant
outcome, and evoking more behaviour that leads to that outcome. 3 showed good motivational effects for IRAP-vetted health related
That is, the statements appeared to alter the value, not the avail- statements even though the self-reported forced choice suggested
ability, of reinforcement. Although this seems to be a more plau- that aesthetic reasons were more important, but Participant
sible account, the current study did not address this question di- 8 failed to show motivational effects given a similar pattern of
rectly and further research is needed to clarify this issue. preference. Participant 10 self-reported that both types of reasons
Furthermore, though the present experiment was designed to were preferred, while the IRAP said neither were, and showed no
determine whether the IRAP could identify statements that were effect for both kinds of statements. Participant 12 showed a similar
likely to function as augmentals, it was not designed to see whe- self-reported pattern, but the IRAP suggested that aesthetic rea-
ther it could do so above and beyond self-report. Self-reported sons would be weak motivators and in fact they were. A much

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 9

person might fleetingly contact the emotions related to looking

good in the eyes of others. Such a transformation of stimulus
functions is a known motivator for actions that might give rise to
contacted consequences (Ju & Hayes, 2008) – it is essentially the
verbal or derived relational extension of direct or conditioned
“reinforcer sampling” (Ayllon & Azrin, 1968; Skinner, 1936). Such
an effect would not be expected if the functions of non-preferred
consequences were made available, and Experiment 2 suggested
that verbal contact with non-preferred consequences did not
function as a motivator.
There are many evidence-based interventions used within
health and fitness research (Martin & Tkachuk, 2000) that might
be augmented by adding verbal motivational tools. IRAP-based
motivational assessment devices might be used in other areas that
present known problems of self-control such as studying, or pro-
crastination. In order to use the IRAP in this way, two main pro-
blems need to be solved.
First, the sampling of domains may need to be broader. Parti-
cipant 10, who failed to show an IRAP effect on either of the two
assessments, reported a long history of exercising on a regular
basis and finding exercise itself to be enjoyable. A wide range of
assessment options within the IRAPs used might have captured
such differences. Participant 8 showed an IRAP effect, but not a
motivative effect. She reported taking the class and participating in
the study primarily because friends were also doing so. Such im-
mediate social reinforcers were also not assessed by the IRAP. It
should also be noted that participants had very limited exposure
to the phrases used and the effects may deteriorate with longer-
term use. Thus for extended application, broader assessment may
be needed.
The second problem is that the IRAP is currently not designed
to identify specific items. Further development of the IRAP will
likely be needed to overcome this problem. Murphy, Mac-
Carthaigh, and Barnes-Holmes (2014) have shown that the tradi-
tional IRAP may still be used at least to examine subgroups of
stimuli, but the issue will potentially exist until it reaches the in-
dividual item level. A new “mixed trial” variant of the IRAP has
been designed to address this issue (Levin, Hayes, & Waltz, 2010).
Although the empirical analysis is very limited, early research
Fig. 4. Participants’ average change from baseline during each condition, for Ex- appears to show some advantages over the traditional IRAP in this
periment 2. regard (Smith, 2013). Regardless of how this problem is solved,
being able to assess this at the individual item level would allow
for greater use of the IRAP in applied settings. That being said, it is
larger study will be needed to settle this remaining point of also worth noting that the changes in actual exercise intensity
comparison but the IRAP appeared to be consistently useful in seen in this study support the robustness of the IRAP as a tool to
identifying the most relevant phrases for participants even as re- identify motivational statements, even given such potential han-
sponses varied across questionnaires for most participants. dicaps as using an older version of the instrument and a rarely
A behavioural explanation of the IRAP is provided by the “Re- used method of identifying its effects.
lational Elaboration and Coherence” model (REC; see Hughes, Many applied problems seem to involve delayed or abstract
Barnes-Holmes, & Vahey, 2012 for a recent description). Relational consequences. The present study shows that human language can
responses differ in their degree of complexity and the current level be used to bring these consequences into the present, motivating
of derivation required, based on the history of the individual and actual effortful behaviour that could give rise to them. Motiva-
context of the task. The IRAP assesses brief and immediate rela- tional augmentals appear to be a highly useful practical tool, given
tional responses (BIRRs) that emerge when derivation is low due only the means needed to identify idiosyncratic reinforcers for key
to a more extended history of responding with regard to the actions of applied importance. It is not yet known if explicit self-
specific relational network and to lower relational complexity. report can also provide those means, and do so equally well, but
These responses may or may not cohere with the more extended the present study suggests that RFT has already provided an im-
and elaborated relational responses (EERRs) that are assessed in plicit means of identifying consequences of importance that can
typical attitudinal measures. For example, a person with BIRRs that lead to the selection of motivative statements in applied situations.
suggest a long and practiced history of relating exercise to looking
good in the eyes of others, may self-edit that response given time to
do so on the grounds that it is vain or superficial. If, during BIRRs,
References
the specific relational responses include consequences of appeti-
tive importance, some of the perceptual functions of those con-
Ayllon, T., & Azrin, N. H. (1968). Reinforcer sampling: A technique for increasing the
sequences can be contacted via a transformation of stimulus behavior of mental patients. Journal of Applied Behavior Analysis, 1, 13–20.
functions through the relational network involved. For example, a Barnes-Holmes, D., Barnes-Holmes, Y., Power, P., Hayden, E., Milne, R., & Stewart, I.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i
10 M.L. Jackson et al. / Journal of Contextual Behavioral Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

(2006). Do you really know what you believe? Developing the Implicit Rela- and Exercise Psychology, 24, 3–32.
tional Assessment Procedure (IRAP) as a direct measure of implicit beliefs. The Hayes, S. C., Barnes-Holmes, D., & Roche, B. (Eds.). (2001). Relational Frame Theory. A
Irish Psychologist, 32(7), 169–177. post-Skinnerian Account of Human Language and Cognition. New York: Kluwer
Barnes-Holmes, D., Murphy, A., Barnes-Holmes, Y., & Stewart, I. (2010). The Implicit Academic/Plenum Press.
Relational Assessment Procedure (IRAP): Exploring the impact of private versus Huang, Y., & Hutchinson, J. W. (2008). Counting every thought: Implicit measures of
public contexts and the response latency criterion on pro-white and anti-black cognitive responses to advertising. Journal of Consumer Research, 35(1), 98–118.
stereotyping among white Irish individuals. The Psychological Record, 60, 57–80. Hughes, S., & Barnes-Holmes, D. (2013). A functional approach to the study of
Barlow, D. H., & Hayes, S. C. (1979). Alternating treatments design: One strategy for implicit cognition: The Implicit Relational Assessment Procedure (IRAP) and the
comparing the effects of two treatments in a single subject. Journal of Applied Relational Eloboration and Coherence (REC) model In: B. Roche, & S. Dymond
Behavior Analysis, 12(2), 199–210. (Eds.), Advances in Relational Frame Theory: Research and application. Reno, NV:
Barnes-Holmes, D., O'Hora, D., Roche, B., Hayes, S. C., Bissett, R. T., & Lyddy, F. Context Press.
(2001). Understanding and verbal regulation In: S. C. Hayes, D. Barnes-Holmes, Hughes, S., Barnes-Holmes, D., & Vahey, N. (2012). Holding on to our functional
& B. Roche (Eds.), Relational Frame Theory: A post-Skinnerian account of human roots when exploring new intellectual islands: A voyage through implicit
language and cognition (pp. 103–117). New York: Plenum Press. cognition research. Journal of Contextual Behavioral Science, 1, 17–38.
Barnes-Holmes, D., Barnes-Holmes, Y., Stewart, I., & Boles, S. (2010). A sketch of the Ju, W. C., & Hayes, S. C. (2008). Verbal establishing stimuli: Testing the motivative
implicit relational assessment procedure (IRAP) and the relational elaboration effects of stimuli in a derived relation with consequences. The Psychological
and coherence (REC) model. The Psychological Record, 60(3), 527–542. Record, 58(3), 339–363.
Barnes-Holmes, Y., Barnes-Holmes, D., Smeets, P. M., & Luciano, C. (2004). A derived Kenward, M., & Roger, J. (1997). Small sample interference for fixed effects from
transfer of mood functions through equivalence relations. Psychological Record, restricted maximum likelihood. Biometrics, 53, 983–997.
54(1), 95–113. Levin, M., Hayes, S. C., & Waltz, T. (2010). Creating an implicit measure of cognition
Berry, T. R., Jones, K. E., McLeod, N. C., & Spence, J. C. (2011). The relationship be- more suited to applied research: A test of the Mixed Trial – Implicit Relational
tween implicit and explicit believability of exercise-related messages and in- Assessment Procedure (MT-IRAP). International Journal of Behavioral Consulta-
tentions. Health Psychology, 30(6), 746–752.
tion and Therapy, 6, 245–262.
Blair, S. N. (2009). Physical inactivity: The biggest public health problem of the 21st
Martin, G. L., & Tkachuk, G. A. (2000). Behavioral sports psychology In: J. Austin, & J.
century. British Journal of Sports Medicine, 43, 1–2.
E. Carr (Eds.), Handbook of applied behavior analysis. Reno, NV: Context Press.
Carpenter, K. M., Martinez, D., Vadhan, N. P., Barnes-Holmes, D., & Nunes, E. V.
McKenna, I. M., Barnes-Holmes, D., Barnes-Holmes, Y., & Stewart, I. (2007). Testing
(2012). Measures of attentional bias and relational responding are associated
the fake-ability of the Implicit Relational Assessment Procedure (IRAP): The
with behavioral treatment outcome for cocaine dependence. The American
first study. International Journal of Psychology and Psychological Therapy, 7(2),
Journal of Drug and Alcohol Abuse, 38(2), 146–154.
253–268.
Cullen, C., Barnes-Holmes, D., Barnes-Holmes, Y., & Stewart, I. (2009). The Implicit
Michael, J. (1982). Distinguishing between discriminative and motivative functions
Relational Assessment Procedure (IRAP) and the malleability of ageist attitudes.
of stimuli. Journal of the Experimental Analysis of Behavior, 37, 149–155.
The Psychological Record, 59, 591–620.
Michael, J. (1993). Establishing operations. The Behavior Analyst, 16(2), 191–206.
Dougher, M. J., Auguston, E., Markham, M. R., Greenway, D. E., & Wulfert, E. (1994).
Milne, R., Barnes-Holmes, D., Barnes-Holmes, Y., & Stewart, I. (2005). Measures of
The transfer of respondent eliciting and extinction functions through stimulus
equivalence functions. Journal of the Experimental Analysis of Behavior, 62, attitudes to autism using the IAT and IRAP. Paper presented at the 31st annual
331–351. convention of the Association for Behavior Analysis. Chicago, USA.
Drake, C. E., Kellum, K. K., Wilson, K. G., Luoma, J. B., Weinstein, J. H., & Adams, C. H. Murphy, C., MacCarthaigh, S., & Barnes-Holmes, D. (2014). Implicit Relational As-
(2010). Examining the Implicit Relational Assessment Procedure: Four pre- sessment Procedure and attractiveness bias: Directionality of bias and influence
liminary studies. The Psychological Record, 60, 81–100. of gender of participants. International Journal of Psychology and Psychological
Gibbons, R. D., Hedeker, D. R., Elkin, I., Waternaux, C., Kraemer, H. C., Greenhouse, J. Therapy, 14(3), 333–351.
B., & Watkins, J. T. (1993). Some conceptual and statistical issues in the analysis Naci, H., & Ioannidis, J. P. A. (2013). Comparative effectiveness of exercise and drug
of longitudinal psychiatric data: Applications to the NIMH Treatment of De- interventions on mortality outcomes: Metaepidemiological study. British
pression Collaborative Research Program dataset. Archives of General Psychiatry, Medical Journal, 347(7929), 5577–5583.
50(9), 739–750. Roche, B., Barnes-Holmes, D., Smeets, P. M., Barnes-Holmes, Y., & McGeady, S.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. K. (1998). Measuring individual (2000). Contextual control over the derived transformation of discriminative
differences in implicit cognition: The Implicit Associations Test. Journal of Per- and sexual arousal functions. The Psychological Record, 50, 267–291.
sonality and Social Psychology, 74, 1464–1480. Skinner, B. F. (1936). The reinforcing effects of a differential stimulus. Journal of
Greenway, D. E., Dougher, M. J., & Wulfert, E. (1996). The transfer of conditioned General Psychology, 14, 263–278.
reinforcement and punishment functions through stimulus equivalence classes. Smith, G. S. (2013). Exploring the predictive utility of the Implicit Relational Assess-
The Psychological Record, 46. ment Procedure (IRAP) with respect to performance in organizations (unpublished
Hagger, M. S. (2012). Advances in motivation in exercise and physical activity In: R. dissertation). Reno: University of Nevada.
M. Ryan (Ed.), The Oxford handbook of human motivation (pp. 479–504). New Valdivia, S., Luciano, C., & Molina, F. J. (2006). Verbal regulation of motivational
York: Oxford University Press. states. The Psychological Record, 56, 577–595.
Hagger, M. S., Chatzisarantis, N. L. D., & Biddle, S. J. J. (2002). A meta-analytic review Winett, R. A. (1998). Developing more effective health-behavior programs: Ana-
of the theories of reasoned action and planned behavior in physical activity: lyzing the epidemiological and biological bases for activity and exercise pro-
Predicitve validity and the contribution of additional variables. Journal of Sport grams. Applied and Preventive Psychology, 7, 209–224.

Please cite this article as: Jackson, M. L., et al. Whatever gets your heart pumping: the impact of implicitly selected reinforcer-focused
statements on exercise intensity. Journal of Contextual Behavioral Science (2015), http://dx.doi.org/10.1016/j.jcbs.2015.11.002i