Journal of Sports Sciences

ISSN: 0264-0414 (Print) 1466-447X (Online) Journal homepage: http://www.tandfonline.com/loi/rjsp20

Motivational priming as a strategy for maximising

exercise outcomes: effects on exercise goals and
engagement

Christian Magaraggia, James Dimmock & Ben Jackson

To cite this article: Christian Magaraggia, James Dimmock & Ben Jackson (2014) Motivational
priming as a strategy for maximising exercise outcomes: effects on exercise goals and
engagement, Journal of Sports Sciences, 32:9, 826-835, DOI: 10.1080/02640414.2013.862841

To link to this article: http://dx.doi.org/10.1080/02640414.2013.862841

Published online: 10 Jan 2014.

Submit your article to this journal

Article views: 820

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rjsp20

Download by: [Vrije Universiteit Amsterdam] Date: 10 November 2017, At: 04:41
 Journal of Sports Sciences, 2014
Vol. 32, No. 9, 826–835, http://dx.doi.org/10.1080/02640414.2013.862841

Motivational priming as a strategy for maximising exercise outcomes:

effects on exercise goals and engagement

CHRISTIAN MAGARAGGIA, JAMES DIMMOCK, & BEN JACKSON

School of Sport Science, Exercise and Health, The University of Western Australia, Perth, Australia

(Accepted 3 November 2013)

Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017

Abstract
Using a scrambled sentence priming protocol, the first aim of this study was to investigate whether differences in novel
exercise-related goals existed between participants primed with motivational or non-motivational material (i.e. autonomous,
controlled or neutral primes). The second aim was to explore whether an indirect effect was present between priming
condition and the goal-related exercise sessions that individuals performed over the week following administration of the
prime. No effects were observed across priming conditions with respect to subjective vitality, goal concordance and the
frequency with which participants planned to exercise. However, autonomy-primed individuals set goals for their exercise
sessions that were significantly longer in intended duration (M = 57.72 min, SD = 37.42) than those set by their counter-
parts in both the controlled-prime condition (M = 44.12 min, SD = 27.78) and the neutral-prime condition
(M = 37.10 min, SD = 20.47). Bootstrapped analyses also revealed a significant indirect relationship between prime and
exercise behaviour, with the autonomy prime predicting longer goal-based exercise sessions, via the effect on the duration of
participants’ intended exercise sessions. These findings highlight the potential influence that priming autonomous motiva-
tion may have on individuals’ exercise aspirations, as well as the way in which primes may indirectly shape exercise
engagement.

Keywords: goal-setting, undergraduate, self-determination theory, goal concordance, intentions, priming

Research over the past few decades has repeatedly
demonstrated that behaviour and evaluations can be
“primed” (e.g. Carver, Ganellen, Froming, &
Chambers, 1983). In other words, the mere percep-
tion of environmental stimuli can influence social
behaviour and judgement without a person’s aware-
ness of this influence. In one seminal priming inves-
tigation, for example, Bargh, Chen, and Burrows
(1996) exposed a treatment group of participants to
elderly related words, such as “old” and “wrinkle”,
and found that individuals in this treatment group
subsequently walked slower at the “completion” of
the experiment than those in a control condition.
These findings suggest that perception of stimuli
associated with social groups can influence our beha-
viour in line with those groups. Aside from stereo-
type-related priming, an abundance of evidence has
indicated that goals and motivation may also be acti-
vated and pursued automatically (e.g. Friedman,
Deci, Elliot, Moller, & Aarts, 2010; Hodgins,
Yacko, & Gottlieb, 2006), and findings by
Levesque and Pelletier (2003) indicate that the prim-
ing of even complex motivational regulations (such
as autonomy and control) can give rise to beha-
vioural outcomes.
Autonomous motivation reflects volitional invol-
vement in an activity for self-determined reasons
(e.g. interest, enjoyment) and is characterised by a
perception of choice over one’s activities (Deci &
Ryan, 2000). Controlled motivation, on the other
hand, is generally associated with pressure that is
either internally or externally imposed (Ryan &
Deci, 2000). That being the case, when individuals
undertake an activity due to autonomous (rather
than controlled) motives, behaviour will be perceived
as an energising and need fulfilling experience (Ryan
& Frederick, 1997; Sheldon, Ryan, & Reis, 1996).
Likewise, the pursuit of goals that are consonant
with internal core values (i.e. self-concordant) can
satisfy basic psychological needs, as well as result in
higher levels of goal attainment and greater well-
being (Sheldon, 2002; Sheldon & Elliot, 1999).
In recent research, our laboratory group (Banting,
Dimmock, & Grove, 2011) primed undergraduate
students in a cycling class with autonomous motiva-
tion, controlled motivation or no particular

Correspondence: James Dimmock, School of Sport Science, Exercise and Health, The University of Western Australia, Perth, Australia.
E-mail: james.dimmock@uwa.edu.au

© 2013 Taylor & Francis


motivation. In each condition, the priming material
involved a task in which participants were required
to unscramble grammatically incorrect sentences
(e.g. “feel are interested I usually” (autonomous);
“feel are pressured I usually” (controlled); “book
we the read top” (neutral)). This task was performed
while participants rode on stationary bicycles at a
university fitness centre, and it was advertised as an
investigation being conducted by the English
Department at the University. Banting et al. discov-
ered that participants primed with autonomous
motivation experienced the cycling as more enjoy-
able relative to controlled-primed participants. Also,
autonomous-primed participants cycled for signifi-
cantly longer, at a greater percentage of their max-
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
imum heart rate, and provided lower ratings of
perceived exertion, relative to their controlled-
primed counterparts. Other research has shown
that these motivational orientations can be primed
via more naturalistic methods (than scrambled sen-
tences), such as by acting in accordance with a par-
ticular orientation (e.g. Friedman et al., 2010).
Consequently, motivational priming, particularly
with respect to autonomous motivation, could be
effective as a strategic tool for use in goal-setting
programmes.
The influence of primes on downstream con-
scious processing
Key findings in the study by Banting et al. (2011)
have already been discussed, but one other finding in
their study is particularly noteworthy. More specifi-
cally, Banting et al. discovered that, relative to the
autonomy-primed and neutral-primed groups, the
participants primed with controlled motivation dis-
played a significantly lower intention to exercise in
the near future. Evidence is therefore emerging to
demonstrate that motivational priming can yield
overt effects, and that these effects might include
changes to conscious intentions.
Although a wealth of studies have documented
non-conscious priming effects (e.g. Bargh et al.,
1996; Levesque & Pelletier, 2003), it is now
acknowledged that research in this field needs to
broaden its scope to include higher-order, or “sec-
ond-generation”, questions (Bargh, 2006). One such
question relates to whether, and how, primes might
interact with more conscious, deliberate ongoing
processes (e.g. one’s stated intentions or goals;
Bargh, 2006). There is evidence to indicate that
goal activation and pursuit can occur without con-
scious intervention (see Bargh, 1990), but there also
remain many instances where consciousness is an
important regulator of behaviour. Overt intentions,
for example, often account for substantial variance in
behaviour, particularly when the temporal disparity
Motivational priming and goal-setting 827
between intention and behaviour is minimised
(Ajzen, 2011). From a conceptual perspective,
there are compelling reasons to suggest that primes
may influence conscious deliberation about an activ-
ity (e.g. goal-setting). In their dual-process model,
for instance, Strack and Deutsch (2004) proposed
that the fast, automatic impulsive system influences
“the differential accessibility of information about
behavioral options or about relevant aspects of the
alternatives” (p. 230), which can bias reflective (i.e.
more conscious) operations. The work by Holland,
Hendriks, and Aarts (2005) has also corroborated
the effects of environmental stimuli on goal activa-
tion and subsequently planned behaviour.
Participants in their treatment group, who had been
exposed to the scent of a cleaning product, were
subsequently more likely to plan activities related to
cleaning than participants in a control group. In a
similar work by Custers and Aarts (2007), partici-
pants who had been primed with the goal of being
“well-groomed” subsequently displayed a greater
likelihood of writing about neatness when asked to
write down their ideal or preferred physical appear-
ance characteristics. Taken together, these findings
indicate that the mere perception of goal-related
stimuli can shape conscious and reflective goal-
related processes, even without participants’ aware-
ness of the influence.
The present study
It is possible that motivational primes offer an oppor-
tunity for shaping individuals’ exercise-related
aspirations. However, researchers are yet to deter-
mine the extent to which abstract motivational
primes can influence one’s overt exercise goals.
Our research group (Banting et al., 2011) provided
preliminary evidence that primes reflecting autono-
mous or controlled motivation can influence exercise
intentions, but, despite these encouraging findings, a
formal goal-setting procedure was not utilised in this
investigation. Moreover, the participants were
primed while they were concurrently undertaking
exercise, and they utilised a generic two-item mea-
sure of intention with responses on Likert scales.
Consequently, information about intended fre-
quency and duration of exercise sessions could not
be gleaned from these items.
Using an undergraduate cohort, the present study
was designed to address these issues. Although the
transition from high school to undergraduate studies
is regularly accompanied by reductions in physical
activity engagement (Pullman et al., 2009), as well as
one’s overall physical health and psychological well-
being (e.g. Bray & Kwan, 2006; Jung, Bray, &
Martin Ginis, 2008), effective strategies for promot-
ing physical activity levels among newly enrolled
 828 C. Magaraggia et al.
undergraduates are lacking (Bray & Born, 2004).
That being the case, first-year undergraduates that
were in their first semester at university were
recruited to participate in the investigation, with the
aim of examining whether goals may support an
indirect pathway between primes and subsequent
exercise behaviour. Relative to participants in the
neutral- and controlled-prime groups, it was
expected that autonomy-primed participants would
create more self-concordant (see Sheldon & Elliot,
1999) exercise goals that were also reflective of
greater intended exercise engagement (i.e. higher
exercise duration and frequency), and that partici-
pants setting goals under this condition would
experience higher subjective vitality (cf. Custers &
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
Aarts, 2007; Holland et al., 2005). It was also
expected that individuals’ exercise goals would pre-
dict their subsequent exercise behaviour (cf. Ajzen,
2011). Overall, then, the aim of this study was to test
if participants primed with autonomous, controlled
or neutral material subsequently set different exer-
cise-related goals, felt differently about those goals
and then subsequently acted on their exercise
intentions.
Method
Participants
A total of 137 first-year undergraduates were
recruited from degree programmes within the
Faculty of Life and Physical Sciences at the authors’
institution. Major fields of study for participants
included physiology, biology, psychology, physics,
kinesiology and chemistry. Two participants were
removed as they reported suspicion that the
scrambled sentence task (i.e. priming procedure)
may have influenced their responses on the goal-
setting task. The remaining sample of 135
(Mage = 18.64, SD = 2.11, nmale = 55, nfemale = 78,
nunknown = 2) completed the first stage of data collec-
tion. Out of the 135 eligible participants who com-
pleted the initial session, 99 participants (nmale = 39,
nfemale = 60), with a mean age of 18.42 years
(SD = 2.00), responded to the follow-up question-
naire at time 2 (i.e. 73.33% retention rate). A one-
way MANOVA was performed to check for potential
background differences between those who did and
did not provide complete data across both time
points, with background (i.e. motivation score, base-
line activity levels and age) and time 1 (i.e. intended
goal frequency and duration, subjective vitality and
goal concordance) indices as dependent variables.
A non-significant multivariate effect emerged
(F7,99 = 1.64, P = .13, η2P = .10, λ = .90), indicating
that “completers” and “dropouts” displayed no dif-
ferences on these variables.
Experimental design and procedures
Having received ethical approval to conduct the
study, students were informed in a lecture that they
could earn course credit in return for their participa-
tion in two studies investigating comprehension abil-
ity (study 1) and the types of exercise goals that
undergraduates set (study 2). Although the studies
were described as being entirely separate, the stu-
dents were informed that the investigations would be
performed back-to-back in an upcoming tutorial. In
this lecture, two separate information sheets and
consent forms were distributed (reinforcing the
bogus cover story that two studies were being con-
ducted). One information sheet explained that the
Faculty of English and Cultural Studies was collect-
ing normative data regarding comprehension ability
among university students. The second information
sheet contained a basic overview of the goal-setting
talk and goal-setting exercise, and outlined that after
the talk students would be invited to set a new
exercise goal for the following week. At this time,
students who consented to participate in the projects
completed a measure assessing their exercise-based
behavioural regulation and initial levels of physical
activity (see the following section).
Experimental session. Approximately two weeks later,
the participants were seated in a lecture room with a
space between each student to minimise possible
distraction and/or any cooperation on the tasks.
The participants were then presented with a 10-min
talk relating to general recommendations for setting
effective goals using information that was obtained
from a recognised undergraduate kinesiology text
(i.e. Weinberg & Gould, 2011). This talk made no
reference to, nor did it include, any material that
may have been associated with either controlled or
autonomous motivational orientations. Importantly,
though, the participants were informed at the end of
this talk that (a) including variety within one’s exer-
cise regime (in terms of activity modality) has been
shown to support exercise maintenance, and (b)
focusing on a novel activity provides exercisers with
an opportunity to incorporate all of the goal-setting
recommendations that were included in the talk.
This was an important component of the design
because novel exercise goals would provide a more
robust test of goal adherence (i.e. avoiding the influ-
ence of other forms of physical activity on results),
and evidence indicates that adolescent students are
likely to be open-minded in relation to setting novel
exercise goals (e.g. Eime, Payne, Casey, & Harvey,
2010). After a short break of 5 min, the students
were invited back into the auditorium and were
asked to begin “study 1”, which consisted of a “com-
prehension” (i.e. scrambled sentence) task. The
 Motivational priming and goal-setting 829

participants received one of three scrambled sen- ended questions about (a) whether they saw a link
tence tasks that had been designed to prime either between studies 1 and 2, and (b) whether they felt
autonomous motivation, controlled motivation or no their judgements or behaviours in study 2 were influ-
particular motivation. The dissemination of these enced by anything that occurred before the study.
tasks across participants was entirely random (i.e.
prior to their arrival, scrambled sentence sheets had
been randomly ordered in a pile, before being dis- Measures
tributed one-by-one among the students upon
Motivational orientation. Participants’ motivational
entry). Respondents were given 5 min to complete
orientations were assessed using the Behavioral
as much of their scrambled sentence task as possible.
Regulation in Exercise Questionnaire-2 (BREQ-2;
As the scrambled sentence materials were col-
Markland & Tobin, 2004; Mullan, Markland, &
lected, the experimenter simultaneously gave the
Ingeldew, 1997), which comprises 19 items that are
participants a goal-setting (“study 2”) survey. To
rated on a five-point response scale anchored at 0
reinforce the separation between prime and depen-
(not true for me) and 4 (very true for me), and is
dent variables, the experimenter thanked the partici-
designed to assess levels of amotivation, external
pants for completing the “first” comprehension
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017

regulation, introjected regulation, identified regula-

study at this time. The participants were then asked
tion and intrinsic motivation. Evidence in support of
to begin the goal-setting (i.e. “second”) study, with
the structural properties of BREQ-2 has been
particular emphasis on setting an exercise goal that
reported previously (Markland & Tobin, 2004),
“they are not currently pursuing but desired to com-
and in the current study, the intrinsic (α = .91),
plete over the course of the following week”. The
identified (α = .78), introjected (α = .81), external
questionnaire (comprising goal-setting task, suspi-
(α = .75) and amotivation (α = .72) subscales all
cion probe and other instruments; see the following
displayed adequate reliability. Scores on these sub-
section) took approximately 15 min to complete,
scales can be used to form a composite relative
after which time the participants were dismissed
autonomy index (RAI) to indicate the extent of indi-
from the laboratory. One week later, the participants
viduals’ autonomous (relative to controlled) regula-
were contacted via email and provided with a written
tion for exercise behaviour (Markland & Ingledew,
copy of their goal (as a reminder), and were then
2007). Prior to computing RAI scores, however, it is
asked to report the number and duration of goal-
necessary to demonstrate that subscale inter-correla-
related exercise sessions that they had performed
tions conform to a simplex pattern (e.g. Markland &
over the week. In a subsequent lecture, all students
Ingledew, 2007) (see Table I for BREQ-2 subscale
were debriefed as to the true nature of the
correlations). In this investigation, intrinsic motiva-
investigation.
tion displayed an acceptable pattern of correlations
with identified, introjected and external regulation.
Priming manipulation. A scrambled sentence task was
Identified regulation was also correlated as antici-
used to administer the prime, and a total of 20
pated with introjected and external regulation, and
sentences were included for each group. The sen-
the association between introjected and external sub-
tences for the autonomous and controlled motiva-
scales supported the expected pattern. Correlations
tion primes were guided by those used in previous
between amotivation scores and external regulation,
investigations (e.g. Banting et al., 2011; Hodgins
introjected regulation, identified regulation and
et al., 2006). To avoid arousing suspicion, a consis-
intrinsic motivation, however, did not conform to
tent number of identical neutral sentences (i.e. one
the forecasted simplex hypothesis. As a result, the
in every four) were included in the autonomous and
amotivation subscale was excluded when computing
controlled priming tasks. In all conditions, only four
RAI scores in this study, and the following RAI
words out of five were required to construct a gram-
calculation formula was used in line with the
matically correct sentence. Examples of the
procedures outlined by Mullan and Markland
scrambled sentences included “options have I two
and” (autonomous, i.e. “I have two options”), “this
do us must we” (controlled, i.e. “we must do this”) Table I. BREQ-2 subscale inter-correlations.
and “book we the read top” (neutral, i.e. “we read
Variable 2 3 4 5
the book”). All scrambled sentence material can be
found in the Appendix. A priming awareness (i.e. Intrinsic motivation .66*** .28** −.35*** −.30**
suspicion) probe was included in order to detect Identified motivation – .53*** −.10* −.53***
whether participants were aware as to the true pur- Introjected regulation – .14 −.08
External regulation – .06
pose of the scrambled sentence task. As per the Amotivation –
suspicion probe used in Dimmock, Jackson, and
Clarke (2013), the participants were asked open- Note: * P < .05, ** P < .01, *** P < .001.
 830 C. Magaraggia et al.
(1997): (–2 × external regulation) + (–1 × introjected
regulation) + (1 × identified regulation) + (2 × intrin-
sic motivation). Higher scores on the RAI indicate a
more autonomous (relative to controlled) regulation.
Exercise goal. The participants were asked to formu-
late and record a new exercise-related goal for the
following week, and to structure their goal around
exercise that they had not undertaken in the previous
month. This instruction ensured that the partici-
pants did not list an exercise goal relating to an
activity in which they may already be engaged. The
participants reported how many exercise sessions
that they intended to complete as part of their exer-
cise goal, and they were also asked to record the
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
intended duration of these sessions (in minutes).
Finally, the modality of the exercise goal (running,
walking and swimming) was also ascertained.
Subjective vitality scale. The subjective vitality scale
(Ryan & Frederick, 1997) was used to assess the
degree to which the participants felt energised/vita-
lised after the goal-setting procedure. The partici-
pants completed an instrument consisting of seven
items rated on a seven-point Likert scale ranging
from 1 (not at all true) to 7 (very true), with items
such as, “At this time, I have energy and spirit” and
“I feel energised right now”. Bostic, Rubio, and
Hood (2000), however, presented evidence that a
six-item measure (which excluded one reverse-
scored item from the original instrument) was super-
ior to the original; therefore, the six-item version was
used for all analyses in the current study. This
six-item measure displayed acceptable reliability
(α = .91) in this investigation.
Goal concordance rating. Goal concordance was mea-
sured by replicating the method used by Sheldon
and Elliot (1999), which involved participants rating
their reasons for pursuing each goal on a nine-point
Likert scale anchored at 1 (not at all for this reason)
and 9 (completely for this reason). Items in this instru-
ment correspond to one of four behavioural regula-
tions, namely, external, introjected, identified and
intrinsic (Sheldon & Elliot, 1999). Items included,
“you pursue this striving because somebody else
wants you to or because the situation demands it”
(external regulation), “you pursue this striving
because you would feel ashamed, guilty or anxious
if you didn’t” (introjected regulation), “you pursue
this goal because you really believe it’s an important
goal to have” (identified regulation) and “you pursue
this striving because of the fun and enjoyment that it
provides you” (intrinsic regulation). A single goal
concordance index was created by weighing the
four behavioural regulations (see Sheldon & Elliot,
1999 for a full description), before summing the
identified regulation and intrinsic motivation scores
and subtracting the introjected and external regula-
tion scores (Sheldon & Elliot, 1999). Accordingly, a
higher goal concordance score reflects goals that are
valued and are pursued primarily due to personal
interest (Sheldon, 2002).
Exercise behaviour. Initial exercise behaviour data
were obtained by asking the participants to report
how many hours of physical activity they engaged in
per week. These baseline levels of exercise were
measured for use in preliminary analyses. One
week after completing the main procedure and set-
ting their new exercise goal, the participants were
contacted via email and asked to report the modality
and frequency (i.e. number) of goal-related exercise
sessions they had completed over the last week, as
well as the duration (i.e. minutes) of each of these
sessions. The participants were asked to only report
the exercise that they had completed which was
directly related to the goal they had set the previous
week and were asked to exclude any exercise that
they had performed which was unrelated to their
goal. Responses to these questions were encouraged
in an open-ended format (i.e. after writing down
their exercise modalities, the participants noted
how many times they undertook each form of exer-
cise and how long they spent in each session). In
line with previous research investigating exercise
behaviour (Ishikawa-Takata, Ohta, & Tanaka,
2003; Juneau et al., 1987), and in order to ensure
concordance between goal- and activity-related
measures, exercise duration (minutes per exercise
session) and frequency (sessions per week) were
included as measures of exercise engagement.
Notably, these measures have been recognised as
important predictors of health outcomes associated
with physical activity (Ballor & Keesey, 1991;
Shephard, 1968).
Results
Preliminary analyses
Descriptive statistics for background and dependent
variables are displayed in Table II, and correlations
between all variables are presented in Table III. A
MANOVA was used to explore whether background
differences existed on participants’ self-reported
motivational orientations (i.e. RAI scores derived
from BREQ-2) and baseline activity levels (i.e. time
spent participating in sport/exercise each week)
according to priming condition. Analyses conducted
on time 1 data revealed that there was no multi-
variate effect (F4,208 = .86, P = .49, η2P = .01, λ =
.97), indicating that there were no background dif-
ferences between those in the three priming
 Motivational priming and goal-setting 831

Table II. Descriptive statistics according to priming condition.

Autonomy Controlled Neutral

M SD M SD M SD

Background variables
RAI score 5.95 2.86 5.36 3.33 5.78 2.59
Physical activity (hours/week) 5.88 3.19 5.86 3.50 5.10 2.46
Dependent variables
Exercise goal frequency 3.74 2.02 4.14 1.76 3.28 1.48
Exercise goal duration 57.72 37.42 44.12 27.78 37.10 20.47
Goal concordance 1.66 1.20 1.47 1.36 1.55 1.22
Subjective vitality 2.91 1.28 3.04 1.33 3.17 1.23
Actual exercise frequency (no. of sessions) 3.05 1.95 2.75 1.40 2.58 2.30
Actual exercise duration (minutes/session) 51.77 29.87 41.31 30.68 35.00 25.71

Notes: Higher RAI scores denote relatively more autonomous (versus controlled) motives for exercise participation. Exercise goal frequency
and duration reflect intended number of exercise sessions and intended duration of each of those sessions (over the coming seven days).
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017

Higher goal concordance score reflects goals that are valued and pursued primarily due to personal interest. Subjective vitality measured 1 to
7, where higher scores denote more positive perceptions. Actual exercise frequency and duration represent the number of goal-related
exercise sessions participated in during the week following administration of the prime, as well as the average duration of each of the sessions.

Table III. Correlation matrix for all background and dependent variables.

Variable 2 3 4 5 6 7 8

RAI score .12 .03 .02 .44*** .34*** .01 .09

Physical activity – .12 .29* .07 .45*** .18 .31**
Exercise goal frequency – −.04 .20* .11 .37*** .08
Exercise goal duration – .06 .15 .17 .64***
Goal concordance – .33*** .09 .17
Subjective vitality – .09 .14
Actual exercise frequency (no. of sessions) – .14
Actual exercise duration (minutes/session) –

Note: * P < .05, ** P < .01, *** P < .001.

conditions (parenthetically, this non-significant mul-
tivariate effect was also apparent when examining
only those who provided time 2 data, F4,176 = .78,
P = .54, η2P = .01, λ = .96). This indicated that
students in the different conditions did not display
differences on exercise-related motivational regula-
tions or the average time they spent participating in
sport/exercise each week. Subsequently, a
MANOVA was also used to determine whether
intended goal duration, intended goal frequency,
goal concordance and subjective vitality differed
according to gender. This analysis was also non-sig-
nificant (F4, 127 = .97, P = .43, η2P = .03, λ = .97),
and as a result, gender was not included as a covari-
ate or independent variable in subsequent analyses.
Main analyses
Exercise goals. A MANOVA was conducted, in which
priming condition was entered as the independent
variable, with intended goal frequency, intended goal
duration, goal concordance and subjective vitality as
the dependent variables. A Bonferroni correction
(i.e. α = .0125) was used to interpret significance at
the univariate level for this analysis. The analysis
revealed a statistically significant multivariate effect
for priming condition (F8,256 = 2.50, P = .012, η2P =
.073, λ = .86). A significant main effect was found
for intended goal duration (F2,131 = 5.54, P = .005,
η2P = .08); however, no significant main effects were
found for intended goal frequency (F2,131 = 2.53,
P = .08, η2P = .04), intended goal concordance
(F2,131 = .27, P = .76, η2P = .004) or subjective
vitality (F2,131 = .42, P = .66, η2P = .006). In relation
to intended goal duration (see Figure 1), autonomy-
primed participants planned to exercise for signifi-
cantly longer within each session (M = 57.72 min,
SD = 37.42) than their controlled- (M = 44.12 min,
SD = 27.78; Mean diff = 13.60, SE = 6.17, P = .03)
and neutral-primed counterparts (M = 37.10 min,
SD = 20.47; Mean diff = 20.63, SE = 6.36, P =
.001). There were no significant intended goal dura-
tion differences between those in the controlled- and
neutral-primed groups (Mean diff = 7.02, SE = 6.39,
P = .27).
 832 C. Magaraggia et al.
60.00
Mean of intended goal duration (minutes)
40.00
20.00
0.00
Autonomy Control
Condition
Figure 1. Means and standard error for goal duration across
priming conditions.
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
Exercise behaviour. To test for the indirect effect
between the prime and subsequent exercise session
duration (via intended goal duration), bootstrap-
ping-based analyses were performed using the
INDIRECT macro for SPSS, developed by
Preacher and Hayes (2008). This procedure tests
each of the individual pathways specified within the
indirect model, as well as providing estimates and
significance levels of total, direct and indirect effects
for the predictor on the outcome. Duration of goal-
based exercise sessions was entered as the dependent
variable, intended goal duration was entered as the
proposed indirect mechanism and priming condition
was entered as a categorical independent variable
(coded autonomy = 1, controlled = 2, neutral = 3).
The bootstrap analysis used 5000 samples to build
non-parametric sampling distributions of the
hypothesised indirect effect, and the indirect effect
is considered to be significant (using α = .05) if the
95% confidence interval excludes zero.
Analyses indicated that the relationship between
priming condition and intended goal duration was
significant (b = –12.10, SE = 4.02, t98 = –3.01, P =
.003), insofar as an autonomous prime was predic-
tive of greater intended session duration. There was
also a significant predictive effect between intended
goal duration and subsequent exercise session dura-
tion (b = .56, SE= .07, t98 = 7.50, P < .001), which
indicated that greater duration intentions were
related to greater actual session duration. Taken
together, these predictors accounted for 39.24% of
the variance in actual exercise session duration
(F2, 95 = 32.38, P < .001). Most importantly, the
bootstrapped total indirect effect from prime to ses-
sion duration (through intended goal duration) was
also significant (estimate = –6.80, SE = 2.31,
95% bias corrected and accelerated confidence inter-
val –12.11, –2.85), and the overall normal theory test
associated with the indirect effect was significant
(z = –2.81, P = .005). Collectively, these findings
indicate that there was a significant indirect prospec-
tive pathway from prime to goal-based exercise ses-
sion duration, via the duration of participants’
planned goal-based sessions.
Discussion
These results indicated that the presentation of an
autonomous (relative to controlled and neutral)
motivational prime may influence the characteristics
of individuals’ exercise goals, at least in terms of
intended goal duration. Importantly, the goals that
Neutral
were set by the participants in the study were pre-
dictive of exercisers’ subsequent behavioural pat-
terns (i.e. the length of exercise sessions carried
out). Findings from the study therefore corroborate
and extend results from other research in which
motivational primes have influenced consciously
planned goals (e.g. Banting et al., 2011). In an
attempt to explain the effects in the present study,
it is noteworthy that scrambled sentence primes for
abstract motivational orientations are “action-neu-
tral” and are not themselves associated with beha-
vioural representations. In other words, unlike
primes that reflect specific goals (e.g. to memorise),
the primes presented in this study reflect broad and
abstract constructs with more distal links to beha-
viour. Consequently, it is likely that the primes in
this study were used to interpret an individual’s
current circumstances, which may have been respon-
sible for the observed responses on goal content (see
Dijksterhuis, Preston, Wegner, & Aarts, 2008).
In terms of goal-related influences, it was interest-
ing that the intended duration, but not the intended
frequency, of exercise sessions differed between
priming conditions. This priming effect is particu-
larly noteworthy given that no differences existed
between the participants in terms of their exercise-
related motivational regulation (i.e. before the
administration of the prime), and the specific influ-
ence on goal duration might reflect the greater per-
ceived malleability of exercise session duration in
comparison to exercise frequency. In fact, goal dura-
tion is often considered the most controllable of the
three components of exercise (i.e. intensity, duration
and frequency) (e.g. Courneya & McAuley, 1994). It
is possible that logistical constraints, such as study or
work commitments, are more likely to impinge on
changes to goal frequency than they are to goal
duration. In other words, autonomy-primed partici-
pants may have been more likely to seek to extend
each of their planned exercise sessions, as opposed
to attempting to schedule additional sessions into
their weekly routine. Somewhat less intuitive,
though, was our finding that priming groups did
not differ in terms of subjective vitality or goal con-
cordance perceptions. One possible explanation for

this lack of effect is that in the current study all
experimental groups may have felt external pressures
to create an exercise goal due to the nature of the
experimental manipulation, and perceived little voli-
tion over the scope of the goal-setting (e.g. Judge,
Bono, Erez, & Locke, 2005). Future work is encour-
aged to overcome this problem by investigating the
impact of primes in self-selected goal-setting
programmes.
Countless studies have supported the notion that
goals influence behaviour, but what is important to
establish is whether goals set under priming condi-
tions retain their motivational force. This study pro-
vided indirect evidence to suggest that prime-
influenced goals still predict behaviour. This finding
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
was particularly noteworthy given that no differences
were apparent between priming groups in terms of
baseline exercise behaviour, and is particularly rele-
vant for this population in light of the decrements in
physical activity and health status that are often
observed during the transition from high school to
undergraduate studies (e.g. Bray & Kwan, 2006).
Although it is not clear whether these findings may
generalise across other populations (e.g. high school
students) and contexts (e.g. community samples),
the indirect prospective effect that emerged provides
encouraging evidence for the role of priming-based
manipulations in promoting exercise-related out-
comes among a first-year undergraduate cohort.
One limitation of the current study, however, was
the self-report nature of the exercise engagement
variable, and in future, researchers are encouraged
to consider the extent to which these primes may
indirectly support physical activity when assessed
using more objective (e.g. accelerometer-based)
methods. Also, an assessment was not made in the
present study about the extent to which the new
goal-related exercise activities operated as substitutes
or as additional activities to participants’ normal
exercise behaviour. Research on naturalistic primes,
another area of inquiry described by Bargh (2006) as
an important “second-generation” issue, also repre-
sents a worthwhile extension to this work. As
opposed to the use of scrambled-sentence tasks,
creating an autonomous or controlled learning envir-
onment in a laboratory prior to a goal-setting activity
might be one “naturalistic” way to investigate this
issue.
Health promotion practitioners often lament the
challenges inherent in changing clients’ exercise
behaviour. Obtaining clients’ commitment to engage
in recommended exercise is often difficult, and even
when this commitment is obtained, clients some-
times do not act upon their intentions. It is encoura-
ging, then, that subtle features of the environment
may be used alongside more traditional cognitive
behavioural approaches to foster more adaptive
Motivational priming and goal-setting 833
exercise goals, which may subsequently align with
desirable behavioural outcomes. To the extent that
this influence is non-conscious, clients are unlikely
to deem the efforts of the practitioner as manipula-
tive or divisive.
References
Ajzen, I. (2011). The theory of planned behavior: Reactions and
reflections. Psychology and Health, 26, 1113–1127.
Ballor, D. L., & Keesey, R. E. (1991). A meta-analysis of the
factors affecting exercise-induced changes in body mass, fat
mass and fat-free mass in males and females. International
Journal of Obesity, 15, 717–726.
Banting, L. K., Dimmock, J. A., & Grove, J. R. (2011). The
impact of automatically activated motivation on exercise-
related outcomes. Journal of Sport & Exercise Psychology, 33,
569–585.
Bargh, J. A. (1990). Auto-motives: Preconscious determinants of
social interaction. In T. E. Higgins & R. M. Sorrentino (Eds.),
Handbook of motivation and cognition: Foundations of social beha-
vior (Vol. 2, pp. 93–130). New York, NY: Guilford Press.
Bargh, J. A. (2006). What have we been priming all these years?
On the development, mechanisms, and ecology of non-con-
scious social behavior. European Journal of Social Psychology,
36, 147–168.
Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of
social behavior: Direct effects of trait construct and stereotype
activation on action. Journal of Personality and Social Psychology,
71, 230–244.
Bostic, T. J., Rubio, M. D., & Hood, M. (2000). A validation of
the Subjective Vitality Scale using structural equation model-
ling. Social Indicators Research, 52, 313–324.
Bray, S. R., & Born, H. A. (2004). Transition to university and
vigorous physical activity: Implications for health and psycho-
logical well-being. Journal of American College Health, 52, 181–
188.
Bray, S. R., & Kwan, M. Y. W. (2006). Physical activity is asso-
ciated with better health and psychological well-being during
transition to university life. Journal of American College Health,
55, 77–82.
Carver, C. S., Ganellen, R. J., Froming, W. J., & Chambers, W.
(1983). Modeling: An analysis in terms of category accessibil-
ity. Journal of Experimental Social Psychology, 19, 403–421.
Courneya, K. S., & McAuley, E. (1994). Are there different
determinants of the frequency, intensity, and duration of phy-
sical activity? Behavioral Medicine, 20, 84–91.
Custers, R., & Aarts, H. (2007). Goal-discrepant situations prime
goal-directed actions if goals are temporarily or chronically
accessible. Personality and Social Psychology Bulletin, 33, 623–
633.
Deci, E. L., & Ryan, R. M. (2000). Intrinsic and extrinsic motiva-
tions: Classic definitions and new directions. Contemporary
Educational Psychology, 25, 54–67.
Dijksterhuis, A., Preston, J., Wegner, D. M., & Aarts, H. (2008).
Effects of subliminal priming of self and god on self-attribution
of authorship for events. Journal of Experimental Social
Psychology, 44, 2–9.
Dimmock, J. A., Jackson, B., & Clarke, D. (2013). “Warm inter-
actions” and “cold streaks”: The influence of temperature word
primes on interpersonal warmth and competence-related out-
comes in a dyadic sport task. Sport, Exercise, & Performance
Psychology, 2, 102–116.
Eime, R. M., Payne, W. R., Casey, M. M., & Harvey, J. T.
(2010). Transition in participation in sport and unstructured
 834 C. Magaraggia et al.
physical activity for rural living adolescent girls. Health
Education Research, 25, 282–293.
Friedman, R., Deci, E. L., Elliot, A. J., Moller, A. C., & Aarts, H.
(2010). Motivational synchronicity: Priming motivational
orientations with observations of others’ behaviors. Motivation
and Emotion, 34, 34–38.
Hodgins, H. S., Yacko, H. A., & Gottlieb, E. (2006). Autonomy
and nondefensiveness. Motivation and Emotion, 30, 283–293.
Holland, R. W., Hendriks, M., & Aarts, H. (2005). Smells like
clean spirit: Non-conscious effects of scent on cognition and
behavior. Psychological Science, 16, 141–154.
Ishikawa-Takata, K., Ohta, T., & Tanaka, H. (2003). How much
exercise is required to reduce blood pressure in essential hyper-
tensives: A dose–response study. American Journal of
Hypertension, 16, 629–633.
Judge, T. A., Bono, J. E., Erez, A., & Locke, E. A. (2005). Core
self-evaluations and job and life satisfaction: The role of self-
concordance and goal attainment. Journal of Applied Psychology,
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
90, 257–268.
Juneau, M., Rogers, F., De Santos, V., Yee, M., Evans, A., Bohn, A.,
& Debusk, R. F. (1987). Effectiveness of self-monitored, home-
based, moderate-intensity exercise training in middle-aged men
and women. The American Journal of Cardiology, 60, 66–70.
Jung, M. E., Bray, S. R., & Martin Ginis, K. A. (2008). Behavior
change and the freshman 15: Tracking physical activity and
dietary patterns in 1st-year university women. Journal of
American College Health, 56, 523–530.
Levesque, C., & Pelletier, L. G. (2003). On the investigation of
primed and chronic autonomous and heteronomous motiva-
tional orientations. Personality and Social Psychology Bulletin, 29,
1571–1584.
Markland, D., & Ingledew, D. K. (2007). The relationships
between body mass and body image and relative autonomy
for exercise among adolescent males and females. Psychology
of Sport and Exercise, 8, 836–853.
Markland, D., & Tobin, V. (2004). A modification to the beha-
vioral regulation in exercise questionnaire to include an assess-
ment of amotivation. Journal of Sport & Exercise Psychology, 26,
191–196.
Mullan, E., & Markland, D. (1997). Variations in self-determina-
tion across the stages of change for exercise in adults.
Motivation and Emotion, 21, 349–362.
Mullan, E., Markland, D., & Ingeldew, D. K. (1997). A graded
conceptualisation of self-determination in the regulation of
exercise behaviour: Development of measures using confirma-
tory factor analytic procedures. Personality and Individual
Differences, 23, 745–752.
Preacher, K., & Hayes, A. (2008). Asymptotic and resampling
strategies for assessing and comparing indirect effects in
multiple mediator models. Behavior Research Methods, 40,
879–891.
Pullman, A. W., Masters, R. C., Zalot, L. C., Carde, L. E.,
Saraiva, M. M., Dam, Y. Y., … Duncan, A. M. (2009).
Effect of transition from high school to university on anthropo-
metric and lifestyle variables in males. Applied Phsyiology,
Nutrition, and Metabolism, 34, 162–171.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and
the facilitation of intrinsic motivation, social development, and
well-being. American Psychologist, 55, 68–78.
Ryan, R. M., & Frederick, C. (1997). On energy, personality, and
health: Subjective vitality as a dynamic reflection of well-being.
Journal of Personality, 65, 529–565.
Sheldon, K. M. (2002). The self-concordance model of healthy
goal striving: When personal goals correctly represent the per-
son. In E. L. Deci & R. M. Ryan (Eds.), Handbook of self-
determination research (pp. 65–86). Rochester, NY: University
of Rochester Press.
Sheldon, K. M., & Elliot, A. J. (1999). Goal striving, need satis-
faction, and longitudinal well-being: The self-concordance
model. Journal of Personality and Social Psychology, 76,
482–497.
Sheldon, K. M., Ryan, R., & Reis, H. T. (1996). What makes for
a good day? Competence and autonomy in the day and in the
person. Personality and Social Psychology Bulletin, 22, 1270–
1279.
Shephard, R. (1968). Intensity, duration and frequency of exercise
as determinants of the response to a training regime. European
Journal of Applied Physiology and Occupational Physiology, 26,
272–278.
Strack, F., & Deutsch, R. (2004). Reflective and impulsive deter-
minants of social behavior. Personality and Social Psychology
Review, 8, 220–247.
Weinberg, R. S., & Gould, D. (2011). Foundations of sport and
exercise psychology (5th ed.). Champaign, IL: Human Kinetics.

Appendix: autonomous, controlled and neutral
scrambled sentences
Autonomous
options have I two and
dollars salad on costs two
feel are choiceful I usually
is to this opportunity my
to we choose so leave
enjoy I freedom my to
very coffee the is hot
in we autonomous often are
have for preference a we
to go and I decided
to our we classes selected
Downloaded by [Vrije Universiteit Amsterdam] at 04:41 10 November 2017
apple was to the delicious
on choice we a have
we today unconstrained were for
can self-regulate to usually I
is to here served lunch
actions and my are independent
at the new computer is
now to I unrestricted am
am I still for self-determined
Motivational priming and goal-setting 835
Controlled Neutral
do we to this must by people walk some now
do I should to homework book we read the top
dollars salad costs two for dollars salad costs two for
to I smile ought and the a brown was dog
for required to I’m study fence they but saw the
work to with obligated I’m two so had he hats
meet we on deadlines must plant I like obvious that
for boss coerced my me was sign a there too
is to here served lunch is to here served lunch
and obey we’re compelled to porch the far white was
compulsory to attendance is our soft indirect is light to
apple was to the delicious apple was to the delicious
giving in to necessary is the walk fish swims slowly
manipulates my to me boss tall is Julia quite but
so behaviour my they restrict pictures and our good were
So coffee the is hot so coffee the is hot
forced by to study I’m I student am a how
at the new computer is at the new computer is
the by limits constrained us are pencils hers the it
very are we pressured that am citizen from a I