Content uploaded by Bianca A. Simonsmeier
Author content
All content in this area was uploaded by Bianca A. Simonsmeier on Jun 23, 2020
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rirs20
International Review of Sport and Exercise Psychology
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rirs20
The effects of imagery interventions in sports: a
meta-analysis
Bianca A. Simonsmeier , Melina Androniea , Susanne Buecker & Cornelia
Frank
To cite this article: Bianca A. Simonsmeier , Melina Androniea , Susanne Buecker & Cornelia
Frank (2020): The effects of imagery interventions in sports: a meta-analysis, International Review
of Sport and Exercise Psychology, DOI: 10.1080/1750984X.2020.1780627
To link to this article: https://doi.org/10.1080/1750984X.2020.1780627
Published online: 23 Jun 2020.
Submit your article to this journal
View related articles
View Crossmark data
The effects of imagery interventions in sports: a meta-analysis
Bianca A. Simonsmeier
a
, Melina Androniea
a
, Susanne Buecker
b
and Cornelia Frank
c
a
Educational Psychology, University of Trier, Trier, Germany;
b
Department of Psychology, Ruhr-University
Bochum, Bochum, Germany;
c
Neurocognition and Action, University of Bielefeld, Bielefeld, Germany
ABSTRACT
Imagery interventions are an established psychological tool to
enhance performance, psychological skills, and injury
rehabilitation. Previous meta-analyses found positive effects of
mental practice on performance, leaving it open whether imagery
can also enhance outcomes other than performance such as
motivational or affective outcomes. We performed a meta-analysis
to extend the current understanding of the effectiveness of
imagery in sports on any sport specific outcome and the
relevance of additional variables potentially moderating the effect.
The overall effect of imagery interventions was medium in
magnitude with d= 0.431 (95% CI [0.298, 0.563]). Imagery
interventions significantly enhanced motor performance,
motivational outcomes, and affective outcomes. Summarized
across all outcomes, imagery combined with physical practice was
more effective than physical practice alone, indicating differential
effects of imagery and physical practice. We found the same
pattern of result for performance outcomes. The effectiveness of
imagery was positively associated with the intensity of the
imagery training. We discuss our results against previous meta-
analyses on mental practice and the background of theoretical
and practical aspects of imagery. Moreover, we lay out directions
for future research by providing a comprehensive overview of
research gaps in the literature on imagery.
ARTICLE HISTORY
Received 6 April 2020
Accepted 5 June 2020
KEYWORDS
Psychological skill; mental
practice; mental strategies;
mental stimulation; review
Imagery is one of the most popular and well-accepted sport psychological strategies to
improve performance, psychological skills, and injury rehabilitation (e.g. Cumming &
Ramsey, 2009; Cumming & Williams, 2013; Guillot & Collet, 2008). It describes
the creation and re-creation of an experience generated from memorial information, involving
quasi-sensorial, quasi-perceptual, and quasi-affective characteristics, that is under the voli-
tional control of the imager, and which may occur in the absence of the real stimulus antece-
dents normally associated with the actual experience. (Morris et al., 2005, p. 19)
Sometimes, imagery and mental practice are used synonymously. The constructs are com-
parable in terms of describing a cognitive process of symbolic rehearsal. However, mental
practice does not necessarily involve imagery but can also refer to other types of mental pro-
cesses or mental preparation including self-talk or relaxation. Different to mental practice,
© 2020 Informa UK Limited, trading as Taylor & Francis Group
CONTACT Bianca A. Simonsmeier simonsm@uni-trier.de
Supplemental data for this article can be accessed at https://doi.org/10.1080/1750984X.2020.1780627
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY
https://doi.org/10.1080/1750984X.2020.1780627
imagery is a specific mental process that can be mentally practiced (Murphy & Martin, 2002)
and that includes the creation and re-creation of an experience of any kind not limited to
motor performance. The present meta-analysis has the aim to extend the current under-
standing of the effectiveness of imagery interventions by investigating the effects of
different types of imagery on various outcomes, employing comprehensive analyses regard-
ing control conditions and comparisons, and using state-of-the-art meta-analytic strategies.
Previous meta-analyses summarized existing evidence from single studies on the effec-
tiveness of mental practice and indicated positive effects on motor performance. Table 1
provides an overview of the main characteristics of the previous meta-analyses regarding
conceptual and methodological aspects. The two earliest published meta-analyses (Feltz &
Landers, 1983; Hinshaw, 1991) investigated the effectiveness of interventions that
included both mental practice and mental preparation which is a more general term
that includes a variety of disparate strategies with the goal to enhance performance (Ber-
tollo et al., 2009; Weinberg et al., 1985). While these meta-analyses provided first evidence
on the effectiveness of mental practice on performance, conclusions about the effective-
ness of mental practice as an individual strategy cannot be drawn. This limitation was
addressed in a later meta-analysis that included studies employing mental practice inter-
ventions only (Driskell et al., 1994). The analysis revealed a medium positive effect of
mental practice on performance when compared to no-treatment control conditions
with d= 0.527 comprising 62 effect sizes from 35 studies. Toth et al. (2020) recently per-
formed a methodological replication of the Driskell et al. (1994) study including 37
studies and 99 effect sizes from the research of the past 25 years. They also found a posi-
tive effect comparable in magnitude of d= 0.419 which was, however, substantially lower
when accounting for publication bias (d= 0.264). Besides demonstrating the effectiveness
of mental practice to enhance performance, the two meta-analyses provide further valu-
able insights into relevant variables moderating the effect leading to useful practical impli-
cations. For example, both meta-analyses highlight the relevance of the duration of the
mental practice session. Results indicated an inverse (Driskell et al., 1994) and inverse U-
shaped relationship (Toth et al., 2020), with both suggesting 20 min sessions to be most
effective.
Limitations and extensions of previous meta-analyses on mental practice
The previous meta-analyses provide valuable theoretical and practical insights on the
effectiveness of mental practice on performance, but do not provide an integration of
the comprehensive existing literature on imagery training thus far. As mental practice
and imagery conceptually differ (Murphy & Martin, 2002), a meta-analysis on the effects
of imagery specifically is still outstanding. The present analysis aims at extending results
of previous meta-analyses in three ways: by implementing different inclusion criteria,
including other potential moderating variables, and applying state of the art statistical
methods which will be discussed in the following.
First, all previous meta-analyses provide evidence for the effectiveness of mental prac-
tice to enhance performance. It is however suggested that imagery is effective to enhance
a variety of outcomes such as psychological skills and not only performance as indicated in
different models on imagery (Cumming & Williams, 2013; Guillot & Collet, 2008; Martin
et al., 1999). This has also been considered in single studies that, for example, investigated
2B. A. SIMONSMEIER ET AL.
Table 1. Characteristics of the previous four meta-analyses on mental practice.
Reference
Standardized
search word
combination
Range
years
included
studies
Explicit inclusion and
exclusion criteria
Accounting for
dependency of
effect sizes
Outlier/
sensitivity
analysis
Tests for
publication bias
Included control
conditions
Included
types of
comparison
Included
outcomes n(k)
d
[95% CI]
or
(SD)
Significant
moderators
Feltz and
Landers
(1983)
N/A 1934–
1981
.The only criterion was
that there be a group
that was given only
mental practice and
that this group have
either pretest scores or
a control group to
which be compared
no no Comparison
effect size
published vs.
unpublished
studies
.Simple control
.Motivational control
(no practice control
group with same
number of
scheduled
experimental
sessions)
.Pre-post
.Post-
post
Performance 60 (146) .48 (.67) .Task type
.Published
vs.
unpublished
studies
Hinshaw
(1991)
N/A 1949–
1987
.One condition mental
practice alone (mental/
physical practice group
combined excluded)
.Adequate control
.Provide the necessary
statistics
no no N/A .Direct
.Motivation (perform
unrelated tasks in
experimental
sessions)
.Pre-post
.Post-
post
Performance 21 (44) .68 (.11) .Type of
mental
practice
.Number of
mental
practice
(minutes)
Driskell
et al.
(1994)
‘mental practice’1934–
1991
.Reported (or allowed
the retrieval of) tests of
performance under a
mental practice
condition in
comparison with a no-
treatment control
condition
no no .Fail safe N .No-contact control
group (e.g. wait-list)
.Equivalent control
group (e.g. non-
treatment activity)
Post-post Performance 35 (62) .527 .Duration of
mental
practice
Toth et al.
(2020)
‘mental practice’
OR ‘imagery’
OR
’visualizationn’
1995–
2018
.Compare the
performance of
participant engaging
in mental practice with
no no .Fail-safe N
.Trim and fill
.Cognitively active
(e.g. mental
arithmetic)
.Cognitively passive
(e.g. reading)
.Pre-post
gains
Performance 37 (99) 0.419
[0.246,
0.597]
.Duration of
mental
practice
.Type of task
(Continued)
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 3
Table 1. Continued.
Reference
Standardized
search word
combination
Range
years
included
studies
Explicit inclusion and
exclusion criteria
Accounting for
dependency of
effect sizes
Outlier/
sensitivity
analysis
Tests for
publication bias
Included control
conditions
Included
types of
comparison
Included
outcomes n(k)
d
[95% CI]
or
(SD)
Significant
moderators
those engaging in NO
practice
.An explicit No-Practice
control group with
which to compare
mental practice group
performance
.Studies in which the
mental practice group
mentally practice the
exact same task as the
one which they were
later expected to
perform
.Minimal cognitive
control (e.g. quiet
rest)
.Type of
imagery
used
4B. A. SIMONSMEIER ET AL.
the effectieness of imagery on other outcomes such as anxiety or self-confidence (e.g.
Callow et al., 2001; Hale & Whitehouse, 1998; Hammond et al., 2012). It is desriable to evalu-
ate the effectiveness of imagery on outcomes other than performance (Wakefield & Smith,
2012) to obtain an overall picture of the effectiveness of imagery interventions in sport. As
the previous meta-analyses did not consider other outcomes than performance, the effect
of imagery on other outcomes is still unclear. To estimate the overall effectiveness of
imagery, we included studies independent of the outcome assessed in the present
meta-analysis. In our classification of outcomes, we followed models on imagery
(Cumming & Williams, 2013; Guillot & Collet, 2008; Martin et al., 1999) and distinguished
between performance outcomes, outcomes on strategies and problem solving, psycho-
logical outcomes including affective and motivational outcomes, rehabilitation outcomes,
and psychological skills.
Second, imagery has emerged as a distinct research field within the domain of sport
psychology and has significantly grown since which led to detailed theories on imagery
(e.g. Cumming & Williams, 2013; Guillot & Collet, 2008; Holmes & Collins, 2001; Martin
et al., 1999). These models include many potential moderators, which have not been con-
sidered in the previous meta-analyses. For example, the Toth et al.’s(2020) meta-analysis
included a moderator analysis considering imagery modality, namely visual, kinesthetic,
and mixed imagery, while the imagery content has not been examined. To classify differ-
ences in the imagery content, five major types have been proposed (Hall et al., 1998;
Paivio, 1985). These are (1) cognitive specific (CS; i.e. images of skills), (2) cognitive
general (CG; i.e. images of strategies), (3) motivational specific (MS, i.e. images of goals),
(4) motivational general-arousal (MG-A, i.e. images of arousal and affect), and (5) motiva-
tional general-mastery (MG-M, i.e. imagery of cognitions including self-confidence and
mental toughness). The cognitive functions are conceptually akin to the term motor
imagery (MI), as both involve the internal representation of motor skills without any cor-
responding body movements occurring (Guillot & Collet, 2008). For the purpose of the
current meta-analysis we included any type of imagery in the analyses and systematically
investigated their effectiveness, respectively.
Third, previous meta-analyses demonstrated the effectiveness of mental practice com-
pared to active and passive control groups without considering differences in the active
control groups (Feltz & Landers, 1983; Hinshaw, 1991)or compared mental practice to
active control groups without any practice components (Driskell et al., 1994; Toth et al.,
2020). In the field of imagery reasearch, most commonly active control groups receive
physical practice (e.g. Guillot et al., 2010; O & Munroe-Chandler, 2008; Shackell & Standing,
2007; Spittle & Kremer, 2010), another sport psychological intervention (e.g. Gordon et al.,
1994; Kim et al., 2017), or another intervention, not related to the sports task (e.g. Robin
et al., 2007). More specifically, imagery combined with physical practice vs. a physical prac-
tice control group is a common form of the application of imagery in research and prac-
tical settings (e.g. Brouziyne & Molinaro, 2005; Frank et al., 2014; C. Wright & Smith, 2009)
but has not been considered in the previous meta-analyses. To extend the findings of the
previous meta-analyses and to estimate the effectiveness of imagery compared to
different control groups, the current meta-analysis included any kind of control group
and systematically investigated variations in the effectiveness of imagery interventions.
Fourth, previous meta-analyses used different types of comparisons to determine the
effects of mental practice potentially leading to different magnitudes in effect sizes.
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 5
While the previous meta-analyses included both, pre–post and post-post comparisons
(Feltz & Landers, 1983; Hinshaw, 1991), or post-post comparisons only (Driskell et al.,
1994), the most recent meta-analysis used pre–post gains to operalize the performance
change due to the mental practice (Toth et al., 2020). Until now, no systematic analysis inves-
tigating differences in the magnitude of effect sizes due to variations in the type of compari-
son exists. Such an analysis is, however, useful to provide benchmark data allowing valid
comparisons across studies and a-priori sample size calculations (Moher et al., 1994; Schwei-
zer & Furley, 2016). We therefore invesigated the effects of imagery interventions taking all
possible types of comparison into account and systematically investigating possible
differences.
Fifth, while all meta-analyses used current statistical methods for their analyses when
they were published, application of modern techniques of meta-analytic aggregation
can further extend the validly and generalizability of their findings. For example, all pre-
vious meta-analyses except for the Toth et al. (2020) meta-analysis ignored the distinction
between correlated designs and independent group designs. As such, conclusions
reached from the meta-analytic results must be questioned (Dunlap et al., 1996).
Further, all previous meta-analyses combined various effect sizes per study. This is proble-
matic as ignoring covariance at the study level can result in biased standard errors and
confidence interval coverage proportions (e.g. Van den Noortgate et al., 2013). Lastly,
none of the previous meta-analyses performed outlier analyses and only few analyses
regarding publication bias, possibly limiting the validity of the results. Novel techniques
of meta-analysis allow the inclusion of more than one effect size per study by statistically
accounting for the dependency (e.g. Hedges et al., 2010) and enabling comprehensive
outlier and publication bias analyses (e.g. Viechtbauer & Cheung, 2010). In the present
meta-analysis, we tried to extend the results found in previous meta-analyses by using
robust methods for aggregating and analyzing the available data meta-analytically.
The present meta-analysis
Imagery is one of the most researched psychological skills within the field of sport psychol-
ogy and is widely accepted in practical settings to enhance performance. Despite the con-
tribution of previous meta-analyses on the effectiveness of mental practice on
performance, the quantitative summary of the existing empirical evidence on the
effects of different types of imagery interventions and the effect on various outcomes
in sports is still outstanding. Further, effects of methodical aspects, for example using
different control groups and types of comparisons to determine the effectiveness of
imagery, are not well understood yet. The present study aims to add to previous meta-ana-
lyses and to fill this gap in the literature on imagery interventions in sports.
We a-priori formulated the following five research questions and hypotheses:
1. Is imagery effective to enhance sport related outcomes and is the effect causal? We
expect to find a positive and causal effect of imagery on sports related outcomes
(Hypothesis 1).
2. Is imagery effective when compared to a passive and an active control condition?
Moreover, is physical practice combined with imagery more effective than physical
practice alone? We assume that the overall effect of imagery interventions compared
6B. A. SIMONSMEIER ET AL.
to different control conditions is positive (Hypothesis 2a) and that physical practice
combined with imagery is more effective than physical practice alone (Hypothesis 2b).
3. Does the magnitude of the effect of imagery vary due to different comparisons
employed? We expect to find a higher effect for pre–post comparisons compared to
effects for post-post and pre–post gains comparisons (Hypothesis 3).
4. Is imagery effective to enhance performance and other sport specific outcomes? We
assume an effectiveness of imagery on various outcomes (Hypothesis 4) (Guillot &
Collet, 2008; Martin et al., 1999).
5. Are different types of imagery (equally) effective? We assume to find positive effects for
all different types of imagery and to obtain no differences between them in their effec-
tiveness (Hypothesis 5).
Besides these main moderators, we included many others for exploratory moderator
analyses retrieved from models on imagery frequently discussed in the literature, for
example, imagery type, imagery perspective, temporal equivalence, the athletes age, or
the athletes expertise (Cumming & Williams, 2013; Guillot & Collet, 2008; Holmes &
Collins, 2001; Martin et al., 1999).
Method
Literature search and study selection
We searched the title, abstract, and keywords of all articles in the literature database Psy-
cINFO and PubMed in January 2018 with the following search string: ((imagery interven-
tion OR imagery training OR mental training OR mental practice) AND (sport* OR athlete*)).
We limited the results to studies with human populations that had been published in a
peer-reviewed journal in English language. We did not limit the search to publication
years. The standardized search with the respective keyword combination and restrictions
provided 290 hits in total. We also identified relevant studies via hand search by looking
through references of published meta-analyses on mental practice and reviews on
imagery (Cooley et al., 2013; Driskell et al., 1994; Feltz & Landers, 1983; Hinshaw, 1991;
Schuster et al., 2011). The hand search identified 62 additional eligible studies. Following,
the systematic search and hand search resulted in 372 relevant hits for this meta-analysis.
Figure 1 summarizes the search process in a PRISMA flow chart (Moher et al., 2009). We
accounted for publication bias by using visual and statistical methods (described in the
section Statistical Analyses below) instead of including unpublished studies. The quality
of unpublished studies is hard to assess, and researchers commonly obtain only a non-
representative set of unpublished studies, which does not increase the quality of the
meta-analytic results (Ferguson & Brannick, 2012; Rothstein & Bushman, 2012).
Inclusion of studies
Inclusion of studies followed four inclusion criteria: (1) The study reported quantitative
non-confounded data on effects of an imagery intervention. The effect could either
pertain to within or between comparisons. Therefore, single case studies, studies with
single subject designs and qualitative studies were excluded. We did not include
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 7
studies/effect sizes comparing one imagery intervention to another imagery intervention
or effect sizes where the treatment group differed in more than the imagery implemen-
tation compared to the control group. Studies combining imagery with other mental strat-
egies were sampled, but only for comparison with studies using pure imagery
Figure 1. Flow chart of the literature search.
8B. A. SIMONSMEIER ET AL.
interventions. Studies were also excluded if they implemented preparatory imagery before
movement execution to extend the validity of the results by combining similar interven-
tions. (2) The imagery content/outcome had to be sports specific. (3) The effects of the
imagery intervention had to pertain to measures assessed before and after the interven-
tion. Studies exclusively assessing online measures during imagery were excluded. (4) The
study had to report original empirical findings (i.e. not a re-analysis of already reported
findings or a review) to avoid double coding of effect sizes.
Coding of studies occurred in two steps. First, all abstracts were screened by the first
author to evaluate relevance for the present study. The third author independently screened
a random sample of 71 studies. The absolute intercoder agreement was 96%. Disagreements
were resolved by discussion. A total of 153 abstracts were classified as eligible for further
inspection of the full texts. Second, the 153 full texts were assessed for eligibility. The
second author coded all full texts; the first and third authors coded half of the full texts,
respectively. Overall, 55 studies were identified as relevant for the present meta-analysis.
The intercoder agreement for the inclusion and exclusion of studies based on inspection
of the full texts was 91%. The overall intercoder agreement for the coding of all moderating
variables and effect sizes was 88%. Disagreements were again resolved by discussion. The
studies included in the meta-analysis are listed in the Supplemental Materials. Whenever
data were missing to compute the effect sizes of interest, we requested this information
from the authors of the paper by email. When the authors did not respond, we were not
able to include the respective study in the meta-analysis.
Data coding
We coded all effect sizes reported in the included studies and did not limit the number of
effects included from each study, because studies using several groups, outcomes, or
measurement points frequently reported several relevant comparisons. To maximize com-
parability across studies, we coded the raw data (i.e. the means and standard deviations of
the outcome measures) instead of the reported effect sizes whenever possible and com-
puted a standardized effect size the same way for each study, as described in the Sup-
plemental Materials. For each included effect, we coded the values of the moderators.
Preparation of effect sizes
For each effect, we computed the effect size Cohen’sdfrom the coded raw data using
syntax (for more details, see the Supplemental Materials). For 61 effect sizes, raw data
was not available. Instead, we included the reported Cohen’sdvalues or the transformed
χ
2
-values, t-values, or F-values. We analyzed three different types of comparisons (pre–
post, post-post, and pre–post gains comparisons) to estimate the overall effectiveness
of imagery interventions which is described in the Supplemental Material in more details.
We performed a bare bones meta-analysis and did not statistically adjust the effect sizes
for study artifacts (Schmidt & Hunter, 2015), as only few studies reported the reliabilities for
the employed measures. As sensitivity analyses, we performed the meta-analysis with and
without attenuation for study artefacts and with small sample correction calculating
Hedge’sginstead of Cohen’sd. The alternative approaches did not meaningfully
change the overall effect sizes and moderating effects.
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 9
Statistical analysis
Before aggregation of data from single studies, we identified outliers using specific outlier
diagnostics for meta-analysis (Viechtbauer & Cheung, 2010). Based on the examination of
studentized deleted residuals, DFFITS values, Cook’s distances, and COVRATIO values (for
more details see Viechtbauer & Cheung, 2010), we removed two effect sizes out of one
study (Afrouzeh et al., 2015).
Most of the included studies provided multiple effect sizes for the effects of an imagery
intervention. Consequently, effect sizes were statistically dependent. We addressed this
problem by using robust variance estimation (RVE, Hedges et al., 2010; Tanner-Smith
et al., 2016; Tanner-Smith & Tipton, 2014). Random effects statistical models were used
for all analysis (Raudenbush, 2009), to address the presumed heterogeneity of effects.
First, we estimated a simple RVE meta-regression model to estimate the overall effect of
imagery interventions. Second, to estimate the variability in the effect size due to modera-
tor variables, we estimated a mixed-effects RVE meta-regression model (for more details,
see the Supplemental Material). We used the robumeta package (Fisher & Tipton, 2014)in
the R statistical environment (R Core Team, 2008) for the analyses.
Meta-analyses can be subject to publication bias, which is a higher likelihood of signifi-
cant results to be published which can in turn bias the overall effect size when aggregating
across studies (Hunter & Schmidt, 2004; Polanin et al., 2016; Rothstein et al., 2005). We
approached this problem by analyzing the symmetry of the distribution around the
mean through visual inspection of funnel plots and computing Egger regressions (Egger
et al., 1997; Lau et al., 2006). Further, we performed trim-and-fill analyses (Duval &
Tweedie, 2000) estimating the number of missing studies that might exist and the
effect that these studies might have had on the overall meta-analytic effect.
Results
Study characteristics
A total of 55 studies reporting 401 effect sizes obtained with 1438 participants were
included in the final meta-analysis. The studies were published between 1960 and 2018
with a median publication year of 2005. The sample size varied between 5 and 106 partici-
pants with a mean of 24.36 participants. The mean age of the participants in the individual
studies ranged from 9 to 40 with an overall mean age of 22 years. A screening of imagery
ability occurred in 44% of the studies. Participants in 2 of the 55 studies had prior experi-
ence with imagery.
Most effect sizes were from randomized controlled trials (62%), followed by controlled
trials (26%), and pre–post comparisons (12%). Studies included outcomes of motor learn-
ing and performance (77%), psychological skills (11%), motivational outcomes (8%),
affective outcomes (3%), and strategies and problem solving (2%).
Overall effectiveness of imagery interventions
Hypothesis 1
Table 2 summarizes all results of the meta-analysis. The meta-analytically obtained overall
effect size based on 401 effect sizes from 55 studies indicated a positive, medium, and
10 B. A. SIMONSMEIER ET AL.
statistically significant effect of imagery with Cohen’sd= 0.431, 95% CI [0.298, 0.563]. The
effect was also significantly positive for randomized controlled trials, indicating a causal
effect of imagery with d= 0.254 (95% CI [0.179, 0.530]). The effects of imagery were
evident in both post-tests and retention tests, indicating lasting changes due to the
intervention.
Hypothesis 2
The effectiveness of imagery was significantly better compared to a passive control
group (no intervention, no physical practice, d= 0.508, 95% CI [0.281, 0.735]) and an
active control group without physical practice (another intervention, no physical prac-
tice, d= 0.484, 95% CI [0.069, 0.898]). Imagery combined with physical practice was
more effective than physical practice alone (d= .0156, 95% CI [0.002, 0.310]). We
repeated this analysis for motor outcomes only, which revealed the same result (d=
0.231, 95% CI [0.060, 0.402]). Overall, the effects of imagery were comparable to the
effects of physical practice (d= 0.125, 95% CI [−0.175, 0.431]). Moderation analyses
did not indicate any differences across the different control conditions with F(7.38) =
0.404, p= .755.
Contributing Factors to the effectiveness of imagery interventions
Hypothesis 3
Regarding hypothesis 3, imagery was effective to enhance motor learning and perform-
ance (d= 0.468, 95% CI [0.303, 0.633)], motivational outcomes (d= 0.348, 95% CI [0.033,
0.663)], and affective outcomes (d= 0.269, 95% CI [0.001, 0.537)]. There were no signifi-
cant differences in the magnitude of the effect depending on the different outcomes (F
(5.89) = 1.32, p= .353). Thus, the results provide evidence for our third hypothesis, indi-
cating the effectiveness of imagery interventions for performance but also psychological
outcomes.
Hypothesis 4
The effectiveness of imagery interventions varied due to the employed comparison (F
(16.90) = 7.53, p= .005). The effect was largest for pre–post comparisons (d= 0.934, 95%
CI [0.571, 1.300)], followed by post-post comparisons (d= 0.570, 95% CI [0.188, 0.953)],
and pre–post gains (d= 0.260, 95% CI [0.122, 0.398)]. We performed the same analyses
holding study characteristics constant. From the set of 32 studies and 278 effect sizes,
results were comparable with the largest effect for pre–post comparisons (d= 0.727,
95% CI [0.548, 0.906)], followed by post-post comparisons (d= 0.339, 95% CI [0.166,
0.511)], and pre–post gains (d= 0.232, 95% CI [0.097, 0.368)]. In summary, effect sizes sig-
nificantly vary in magnitude due to the employed comparison, as expected in our fourth
hypothesis.
Hypothesis 5
Few studies reported their employed imagery type, leading to insufficient power to deter-
mine effect sizes and moderation analyses. Whenever stated in the single studies, CS
imagery was implemented most commonly, followed by MG-M imagery, MG-A imagery,
and other types of imagery. Cognitive specific imagery was effective to enhance sport
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 11
Table 2. Number of studies (j), Number of effect sizes (k), Effect size (d), 95% Confidence interval,
measure of heterogeneity τ
2
, and significance of the moderator analyses.
Coding Options j k d 95% CI τ
2
Sign. moderator
Methodological study characteristics
Study Design *
Randomized controlled trial 30 249 0.254 [0.179, 0.530] .202 ref
Controlled trial 16 104 0.347 [0.131, 0.564] .520 ns
Pre-Post 10 48 0.934 [0.571, 1.300] .164 **
Control group ns
No intervention, no physical practice control group 17 51 0.508 [0.281, 0.735] ns
No intervention, physical practice control group 25 153 0.138 [−0.045, 0.322] ref
Another intervention, no physical practice control group 9 42 0.484 [0.069, 0.898] ns
Another intervention, physical practice control group 10 107 0.488 [0.125, 0.851] ns
Comparison **
Pre-post gains 33 283 0.260 [0.122, 0.298] 0.267 ref
Post-post 11 61 0.570 [0.188, 0.953] 0.336 ns
Pre-post 10 48 0.934 [0.571, 1.300] 0.164 **
Retention Test ns
Yes 8 26 0.429 [0.252, 0.572] .236 ref
No 52 368 0.441 [0.252, 0.572] .337 ns
Outcome Category ns
Motor learning and performance 48 304 0.416 [0.272, 0.559] .273 ns
Strategies and problem solving 2 6 0.032 –– –
Affective 6 13 0.269 [0.001, 0.537] .161 ref
Motivational 7 30 0.351 [0.003, 0.699] .151 ns
Psychological skills 6 44 0.583 [−0.362, 1.530] 1.013 *
Imagery intervention characteristics
Imagery Type –
CS 24 194 0.304 [0.135, 0.473] .186
MG-A 2 12 0.476 ––
MG-M 3 15 0.907 ––
Other 2 16 -.0148 ––
Sports ns
Archery 2 7 0.005 ––
Basketball 6 33 0.524 [0.026, 1.020] .104 ns
Cricket 2 64 0.190 ––
Darts 3 10 0.309 ––
Field hockey 2 5 2.170 ––
Figure Skating 2 33 0.297 ––
Fitness 8 55 0.161 [−0.219, 0.541] .219 ref
Golf 8 64 0.557 [0.144, 0.970] .322 ns
Gymnastics 3 14 0.455 ––
Soccer 3 21 0.265 ––
Introduction ns
Yes 23 144 0.491 [0.239, 0.742] .264 ns
No 28 244 0.445 [0.289, 0.600] .334 ref
Duration [days] cont. 53 386 ns
Duration [minutes/session] cont. 33 284 ns
Number of sessions cont. 49 374 **
Number of trials cont. 28 159 ns
Modality of instruction ns
Read aloud 5 33 .201 ––
Audio tape 7 80 0.418 [−0.116, 0.952] . 211 ref
Inner self-talk 35 252 0.501 [0.303, 0.698] .391 ns
Various 3 17 0.433 ––
Setting ns
Practice 20 187 0.242 [0.080, 0.403] .171 ref
Outside of practice 16 98 0.611 [0.231, 0.992] .543 ns
Senses –
Visual 5 14 0.450 ––
Kinesthetic 5 76 0.232 ––
Combination 28 163 0.507 [0.296, 0.718] .277
(Continued)
12 B. A. SIMONSMEIER ET AL.
specific outcomes with d= 0.304 (95% CI [0.135, 0.473)]. Due to the lack of studies and
resulting limited power, we were not able to perform any statistical analyses regarding
our fifth hypothesis.
Exploratory moderators
Of all included moderators, the only significant moderator was the number of sessions of
the imagery training. The effectiveness was significantly higher the more sessions the
implementation included (F(18) = 11, p= .004). None of the other moderators reached sig-
nificance, demonstrating a robust effect of imagery across different age groups, imagery
specifics, and settings.
Checks for publication bias
Visual inspection of the funnel plot (see Figure 2) for the study level (i.e. the average effect
size of each study) and effect size level (i.e. all 401 effect sizes) and Egger regressions for
random effects (Egger et al., 1997) did not indicate significant asymmetry on study level
Table 2. Continued.
Coding Options j k d 95% CI τ
2
Sign. moderator
Script ns
Given 23 183 0.440 [0.238, 0.642] .231 ref
Individual 20 150 0.538 [0.290, 0.787] .284 ns
Timing (Temporal equivalence) –
Yes 14 81 0.568 [0.214, 0.922] .350
No 3 67 0.045 ––
Mixed 3 9 0.439 ––
Stimulus-and-response propositions ns
yes 14 66 0.696 [0.286, 1.110] .414 ns
no 32 257 0.396 [0.235, 0.558] .306 ref
Perspective ns
Internal 28 175 0.428 [0.251, 0.605] .185 ref
External 5 42 0.205 ––
Internal and External 3 65 0.392 ––
Individual 7 48 0.479 [−0.114, 1.070] .398 ns
Athlete characteristics
Age cont. 40 281 ns
Age group ns
Children and Adolescents (7-18) 15 128 0.577 [0.313, 0.840] 0.260 ns
Adults (18 and older) 39 267 0.412 [0.252, 0.572] 0.311 ref
Mixed 2 6 −0.083 –– –
Sports Expertise ns
Novice 26 204 0.268 [0.155, 0.580] 0.322 ref
Recreational 10 69 0.552 [0.299, 0.806] 0.207 ns
Competitive 10 81 0.555 [0.185, 0.926] 0.390 ns
Professional 9 31 0.390 [0.030, 0.751] 0.239 ns
Various 3 16 ––––
Previous imagery experience –
Yes 2 12 0.268 ––
No 48 296 0.449 [0.311, 0.587] 4.350
Imagery ability –
Low to moderate 2 4 0.118 ––
Moderate to High 12 64 0.928 [0.551, 1.310] .364
Note: –: insufficient number of data points for the analysis; ref: reference category; ns: not significant, *p< .05, **p< .01.
Moderators without levels were used as continuous predictors in meta-regressions; all moderators with only one study
were omitted from this table.
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 13
and effect size level. Trim and fill (Duval & Tweedie, 2000) did not indicate any missing
studies, keeping the results of the meta-analysis unchanged. The results suggest that
our findings are unlikely distorted by publication bias.
Discussion
Effectiveness of imagery interventions
The main goal of the meta-analysis was to extend previous meta-analyses on mental prac-
tice by examining the overall effects of imagery on sport specific outcomes, its generaliz-
ability, and identifying relevant moderating variables. We summarize the results of the
meta-analysis in five key findings. First, imagery is effective to enhance not only perform-
ance but also other sport specific outcomes. Second, imagery significantly enhances the
effects of physical practice. Third, the magnitude of effect sizes vary across different
types of comparisons. Fourth, imagery effects are dosage specific. Fifth, there still exist
many gaps in the literature on imagery. While the different theories and models include
a variety of variables proposed to be relevant for successful imagery interventions, empiri-
cal evidence of the relevance of these variables is lacking.
Figure 2. Funnel plots of the inverse standard error and the effect size Cohens’s d for the effect size
level (left) and the study level (right).
14 B. A. SIMONSMEIER ET AL.
Key finding 1
As expected, imagery was effective to enhance sport specific outcomes with an overall
effect size of d= 0.431. The obtained effect size summarizing various sport specific out-
comes is comparable to previous meta-analyses on mental practice focusing on perform-
ance outcomes, which found effect sizes of d= 0.527 (Driskell et al., 1994) and d= 0.419
(Toth et al., 2020). It is also comparable to the effects from other meta-analyses investi-
gating effects of psychological interventions on sports performance, such as goal
setting (Kyllo & Landers, 1995) or self-talk interventions (Hatzigeorgiadis et al., 2011).
The effect of imagery is causal rather than correlation, as the effect was still present in
randomized controlled trials holding third variables constant. The imagery was signifi-
cantly more effective compared to a passive control group and an active control group.
Results suggest a long-term effect as the effectiveness was demonstrated in retention
tests. It was further evident across a variety of different participant, intervention, and
outcome characteristics, indicating a robust effect of imagery across different settings.
Key finding 2
Imagery combined with physical practice was more effective than physical practice alone,
which follows the assumption that imagery enhances the effects of physical practice (e.g.
Hall et al., 1992). From a theoretical perspective, imagery and physical practice share
similar neural substrate, although corresponding neural networks are not totally overlap-
ping (e.g. Decety et al., 1994; Gerardin et al., 2000; Guillot et al., 2008). More recently, it has
been discussed that combined mental and physical practice is superior to physical practice
alone as there might be differential influences of mental practice and physical practice
(e.g. Frank et al., 2016; Guillot et al., 2013). Another explanation might be that the com-
bined practice group had more practice trials as compared to the physical practice
group. However, when analyzing the effectiveness of studies with equal practice trials
(e.g. imagery plus physical practice vs. physical practice plus a filler task), the effect of
imagery was still positive and significant, both on motor tasks and other tasks. Therefore,
the results of the current meta-analysis also indicate differential influences of imagery and
physical practice. Furthermore, the differential effects are likely to be found for both motor
performance and other sport specific outcomes such as psychological outcomes.
Key finding 3
We demonstrate that the magnitude of effect size for imagery interventions differs due to
the type of comparison. Our results highlight the relevance to account for different com-
parisons, as pre–post effects significantly differ from post-post effects and pre–post gains.
The last comparison is most appropriate for intervention studies that focus on learning or
improvement in specific variables as it represents the change due to an intervention (e.g.
Hake, 1998; Nissen et al., 2018; R. H. Williams & Zimmerman, 1996). Studies aiming to inves-
tigate change in specific variables due to an imagery intervention should therefore report
gain scores whenever possible. Many studies reported the relevant information for the
computation of gains scores, but did not report the effect size respectively. One reason
why studies do not report gain scores might be that gain scores were once thought to
have undesirable statistical characteristics, for example, chronically low reliabilities.
Newer studies, however, show that these conclusions are based on unrealistic statistical
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 15
assumptions and that gain scores can have acceptable validities and reliabilities under
more realistic assumptions (Maris, 1998; May & Hittner, 2010; Zimmermann & Williams,
1998).
Key finding 4
Effects of imagery are dosage specific. Our results suggest that the more sessions
employed, the more effective the imagery intervention. This result supports the results
from previous meta-analyses and reviews, also indicating dosage specificeffects of
mental practice (Driskell et al., 1994; Hinshaw, 1991; Toth et al., 2020) and imagery inter-
ventions (Cooley et al., 2013). The results further support expertise theory, suggesting that
the time an individual is engaged in deliberate practice activities is monotonically related
to performance enhancements (Ericsson et al., 1993). Deliberate practice is any type of
practice activity with the aim to improve performance. Imagery has been described as
one form of deliberate practice in sport (Cumming & Hall, 2002). Given that imagery is
more effective the more sessions employed, our findings provide evidence for the poten-
tial of imagery as one form of deliberate practice. However, the interventions of the single
studies were designed to find positive and high effects of imagery with the majority of the
studies following best practice strategies. As such, it is still unclear whether too much
imagery may lose its effects.
Key finding 5
There still exist many gaps in research on imagery, resulting in an underrepresentation of
some moderators in our analyses. While theories and models include variables considered
relevant for imagery to be successful, their effect remains to be determined. On a descrip-
tive level, we found great differences in the magnitude of effect size regarding, for
example, imagery type, the participants’imagery ability or skill level, the employed per-
spective, or temporal equivalence of the imagery. However, we were not able to test
these differences statistically due to a small number of studies within one category.
One reason may be the complex nature of imagery and the large amount of differences
in the implementation. In turn, it needs a great amount of studies to test differences in
the effectiveness caused by the specifics of imagery and implementation systematically.
To understand the impact of characteristics of the participants, the specifics of imagery
and the implementation strategies, more empirical investigations are needed that
thoroughly test the variety of variables of models on imagery (Cumming & Williams,
2013; Guillot & Collet, 2008; Holmes & Collins, 2001).
Directions for future research
In light of the current meta-analysis, at least three suggestions for future research can be
proposed. First, specifics of imagery greatly varied across studies, resulting in few studies
within each moderator category. To fully understand the effectiveness of imagery and
under which condition imagery is most effective, more research is needed that tests
assumptions of current models on imagery. The summary of previous studies in our
meta-analysis gives a comprehensive overview of remaining gaps in literature. For
example, future studies could investigate the effects of imagery other than cognitive
specific imagery (e.g. Martin et al., 1999; Munroe-Chandler et al., 2012)ordifferences in
16 B. A. SIMONSMEIER ET AL.
the effectiveness of imagery due to participant characteristics, such as imagery ability (e.g.
S. E. Williams & Cumming, 2011) and expertise (Guillot et al., 2008; Simonsmeier et al.,
2018). Further, variations in the implementation, such as temporal equivalence or
imagery perspective (e.g. Cumming & Ste-Marie, 2001; Holmes & Collins, 2001) can be con-
sidered in future studies. Replication of existing results and extension of empirical evi-
dence is necessary to test whether the results generalize to other settings and other
populations.
Second, most of the outcomes related to motor performance. Only a few studies inves-
tigated the effects of imagery on psychological skills, such as imagery use (e.g. Cumming &
Ste-Marie, 2001; Munroe-Chandler et al., 2005) and imagery ability (e.g. Cumming & Ste-
Marie, 2001; D. J. Wright et al., 2015), motivational outcomes (e.g. Martin & Hall, 1995;
Ramsey et al., 2010), or affective outcomes (e.g. Mellalieu et al., 2009; Ramsey et al.,
2010). Future research may extend the understanding of the effectiveness of imagery
by investigating effects of imagery on other outcomes than performance.
Third, the quality of future studies and meta-analyses may be enhanced. This can be
achieved by critically deciding for a specific research design, control groups, comparison,
and reporting the procedure of an intervention more precisely (e.g. Goginsky & Collins,
1996). We had to exclude 90 studies for methodological reasons or reporting issues,
such as missing data (e.g. missing means or standard deviations) or confounding variables
(e.g. combination of imagery with other mental strategies). Researchers should conscien-
tiously decide on how they report characteristics of their studies to ensure replicability. In
our case, most imagery interventions were described incomplete, resulting in fewer effect
sizes that could be included in moderator analyses as compared to the amount of effect
sizes included in the general analysis.
Generalizability of the findings and conclusion
The medium positive effect of imagery in sport specific outcomes reported in the present
meta-analysis can be interpreted as non-confounded effects of imagery, as studies were
only included when the intervention and control condition only differed in the implemen-
tation of imagery (e.g. imagery compared to no intervention) and nothing else (e.g. imagery
combined with physical practice compared to no intervention). As the effect of imagery was
still significant for randomized controlled trials, possible influences of third variables can be
ruled out. The effects were robust as compared to passive control groups and active control
groups implementing another intervention. Further, effects of imagery were evident in both
post-tests and retention tests, indicating lasting changes due to the intervention. In our
analysis, we accounted for publication bias and the dependencies of effect sizes, and had
sufficient coding reliability to increase the validity of our findings.
Due to the great variety in the implementation of imagery interventions, some modera-
tors were underrepresented. For example, there is little evidence on the effects of imagery
for participants with low imagery ability, professional athletes, or participants with imagery
experience. Similarly, only a few studies investigated the effects of imagery types other
than cognitive specific imagery. Furthermore, some specific characteristics of the
imagery implementation were considered only in a few studies, for example, variations
in timing, the perspective, or the sensory modalities included. This resulted in limited
power for some analyses.
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 17
Overall, imagery was confirmed to be an effective strategy for enhancing various out-
comes in sports. Furthermore, we identified moderators that could further explain the
effects of imagery, enhance our knowledge regarding the effectiveness of imagery inter-
ventions, and provide directions for future research. The medium effect of imagery in
sport, as determined by the present meta-analysis, encourages the promotion and use
of imagery by athletes, sport educators, coaches, and sport psychologists. The obtained
data in this meta-analysis may serve as benchmark data for future studies with varying
designs (e.g. within or between comparisons) and comparisons (e.g. post-post compari-
sons or pre–post gain comparisons) and can be used for a-priori power analyses. As
such, we hope that this meta-analysis provides new and valuable insights into the
impact of imagery on various outcomes in sports.
Disclosure statement
No potential conflict of interest was reported by the author(s).
ORCID
Bianca A. Simonsmeier http://orcid.org/0000-0002-2269-4838
References
Afrouzeh, M., Sohrabi, E., Haghkhan, A., Rowshani, F., & Goharrokhi, S. (2015). Effectiveness of PETTLEP
imagery on performance of passing skill in volleyball. The Journal of Sports Medicine And Physical
Fitness,55(1-2), 30–36.
Bertollo, M., Saltarelli, B., & Robazza, C. (2009). Mental preparation strategies of elite modern pentath-
letes. Psychology of Sport and Exercise,10(2), 244–254. https://doi.org/10.1016/j.psychsport.2008.
09.003
Brouziyne, M., & Molinaro, C. (2005). Mental imagery combined with physical practice of approach
shots for golf beginners. Perceptual and Motor Skills,101(1), 203–211. https://doi.org/10.2466/
PMS.101.5.203-211
Callow, N., Hardy, L., & Hall, C. (2001). The effects of a motivational general-mastery imagery interven-
tion on the sport confidence of high-level badminton players. Research Quarterly for Exercise and
Sport,72(4), 389–400. https://doi.org/10.1080/02701367.2001.10608975
Cooley, S. J., Williams, S. E., Burns, V. E., & Cumming, J. (2013). Methodological variations in guided
imagery interventions using movement imagery scripts in sport: A systematic review. Journal of
Imagery Research in Sport and Physical Activity,8(1), 13–34. https://doi.org/10.1515/jirspa-2012-
0005
Cumming, J., & Hall, C. (2002). Deliberate imagery practice: The development of imagery skills in com-
petitive athletes. Journal of Sports Sciences,20(2), 137–145. https://doi.org/10.1080/02640410
2317200846
Cumming, J., & Ramsey, R. (2009). Imagery interventions in sport. In S. Mellalieu & S. Hanton (Eds.),
Advances in applied sport psychology: A review (pp. 5–36). Routledge.
Cumming, J., & Ste-Marie, D. M. (2001). The cognitive and motivational effects of imagery training: A
matter of perspective. The Sport Psychologist,15(3), 276–288. https://doi.org/10.1123/tsp.15.3.276
Cumming, J., & Williams, S. E. (2013). Introducing the revised applied model of deliberate imagery use
for sport, dance, exercise, and rehabilitation. Movement & Sport Sciences,82(4), 69–81. https://doi.
org/10.1051/sm/2013098
18 B. A. SIMONSMEIER ET AL.
Decety, J., Perani, D., Jeannerod, M., Bettinardi, V., Tadary, B., Woods, R., Mazziotta, J. C,, & Fazio,, F.
(1994). Mapping motor representations with positron emission tomography. Nature,371(6498),
600–602. https://doi.org/10.1038/371600a0
Driskell, J. E., Copper, C., & Moran, A. (1994). Does mental practice enhance performance? Journal of
Applied Psychology,79(4), 481–492. https://doi.org/10.1037/0021-9010.79.4.481
Dunlap, W. P., Cortina, J. M., Vaslow, J. B., & Burke, M. J. (1996). Meta-analysis of experiments with
matched groups or repeated measures designs. Psychological Methods,1(2), 170–177. https://
doi.org/10.1037/1082-989X.1.2.170
Duval, S., & Tweedie, R. (2000). A nonparametric “trim and fill”method of accounting for publication
bias in meta-analysis. Journal of the American Statistical Association,95(449), 89–98. https://doi.org/
10.2307/2669529
Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple,
graphical test. British Medical Journal,315(7109), 629–634. https://doi.org/10.1136/bmj.315.7109.
629
Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acqui-
sition of expert performance. Psychological Review,100(3), 363–406. https://doi.org/10.1037//
0033-295X.100.3.363
Feltz, D. L., & Landers, D. M. (1983). The effects of mental practice on motor skill learning and perform-
ance: A meta-analyses. Journal of Sport Psychology,5(1), 25–57. https://doi.org/10.1123/jsp.5.1.25
Ferguson, C. J., & Brannick, M. T. (2012). Publication bias in psychological science: Prevalence,
methods for identifying and controlling, and implications for the use of meta-analyses.
Psychological Methods,17(1), 120–128. https://doi.org/10.1037/a0024445
Fisher, Z., & Tipton, E. (2014). Robumeta: Robust variance meta-regression.http://cran.rproject.org/
web/packages/robumeta/index.html
Frank, Cornelia, Land, William M., Popp, Carmen, Schack, Thomas, & Urgesi, Cosimo. (2014). Mental
Representation and Mental Practice: Experimental Investigation on the Functional Links
between Motor Memory and Motor Imagery. PLoS ONE,9(4), e95175. doi:10.1371/journal.pone.
0095175
Frank, C., Land, W. M., & Schack, T. (2016). Perceptual-cognitive changes during motor learning: The
influence of mental and physical practice on mental representation, gaze behavior, and perform-
ance of a complex action. Frontiers in Psychology,6. Article 1981. https://doi.org/10.3389/fpsyg.
2015.01981
Gerardin, E., Sirigu, A., Lehericy, S., Poline, J. B., Gaymard, B., Marsault, C., Agid, Y., & Le Bihan, D. (2000).
Partially overlapping neural networks for real and imagined hand movements. Cerebral Cortex,10
(11), 1093–1104. https://doi.org/10:1093–1104
Goginsky, A. M., & Collins, D. (1996). Research design and mental practice. Journal of Sports Sciences,
14(5), 381–392. https://doi.org/10.1080/02640419608727725
Gordon, S., Weinberg, R., & Jackson, A. (1994). Effect of internal and external imagery on cricket per-
formance. Journal of Sport Behavior,17(1), 60–75.
Guillot, A., & Collet, C. (2008). Construction of the motor imagery integrative model in sport: A review
and theoretical investigation of motor imagery use. International Review of Sport and Exercise
Psychology,1(1), 31–44. https://doi.org/10.1080/17509840701823139
Guillot, A., Collet, C., Nguyen, V. A., Malouin, F., Richards, C., & Doyon, J. (2008). Functional neuroana-
tomical networks associated with expertise in motor imagery. NeuroImage,41(4), 1471–1483.
https://doi.org/10.1016/j.neuroimage.2008.03.042
Guillot, A., Moschberger, K., & Collet, C. (2013). Coupling movement with imagery as a new perspec-
tive for motor imagery practice. Behavioral and Brain Functions,9(1), 8. https://doi.org/10.1186/
1744-9081-9-8
Guillot, A., Tolleron, C., & Collet, C. (2010). Does motor imagery enhance stretching and flexibility?
Journal of Sports Sciences,28(3), 291–298. https://doi.org/10.1080/02640410903473828
Hake, R. R. (1998). Interactive-engagement versus traditional methods: A six-thousand-student
survey of mechanics test data for introductory physics courses. American Journal of Physics,66
(1), 64–74. https://doi.org/10.1119/1.18809
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 19
Hale, B. D., & Whitehouse, A. (1998). The effects of imagery-manipulated appraisal on intensity and
direction of competitive anxiety. The Sport Psychologist,12(1), 40–51. https://doi.org/10.1123/tsp.
12.1.40
Hall, C., Buckolz, E., & Fishburne, G. J. (1992). Imagery and the acquisition of motor skills. Canadian
Journal of Sport Sciences,17(1), 19–27.
Hall, C., Mack, D. E., Paivio, A., & Hausenblas, H. A. (1998). Imagery use by athletes: Development of the
sport imagery questionnaire. International Journal of Sport Psychology,29(1), 73–89.
Hammond, T., Gregg, M., Hrycaiko, D., Mactavish, J., & Leslie-Toogood, A. (2012). The effects of a moti-
vational general-mastery imagery intervention on the imagery ability and sport confidence of
inter-collegiate golfers. Journal of Imagery Research in Sport and Physical Activity,7(1), Article 3.
https://doi.org/10.1515/1932-0191.1066
Hatzigeorgiadis, A., Zourbanos, N., Galanis, E., & Theodorakis, Y. (2011). Self-talk and sports perform-
ance: A meta-analysis. Perspectives on Psychological Science,6(4), 348–356. https://doi.org/10.1177/
1745691611413136
Hedges, L. V., Tipton, E., & Johnson, M. C. (2010). Robust variance estimation in meta-regression with
dependent effect size estimates. Research Synthesis Methods,1(2), 39–65. https://doi.org/10.1002/
jrsm.5
Hinshaw, K. E. (1991). The effects of mental practice on motor skill performance: Critical evaluation
and meta-analysis. Imagination, Cognition and Personality,11(1), 3–35. https://doi.org/10.2190/
X9BA-KJ68-07AN-QMJ8
Holmes, P. S., & Collins, D. J. (2001). The PETTLEP approach to motor imagery: A functional equival-
ence model for sport psychology. Journal of Applied Sport Psychology,13(1), 60–83. https://doi.org/
10.1080/10413200109339004
Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error and bias in research
findings. Sage Publication Inc.
Kim, T., Frank, C., & Schack, T. (2017). A systematic investigation of the effect of action observation
training and motor imagery training on the development of mental representation structure
and skill performance. Frontier in Human Neuroscience,11, 499. https://doi.org/10.3389/fnhum.
2017.00499
Kyllo, L. B., & Landers, D. M. (1995). Goal setting in sport and exercise: A research synthesis to resolve
the controversy. Journal of Sport and Exercise Psychology,17(2), 117–137. https://doi.org/10.1123/
jsep.17.2.117
Lau, J., Ionnidis, J. P. A., Terrin, N., Schmidt, C. H., & Olkin, I. (2006). The case of the misleading funnel
plot. British Medical Journal,333(7568), 597–600. https://doi.org/10.1136/bmj.333.7568.597
Maris, E. (1998). Covariance adjustment versus gain scores—revisited.. Psychological Methods,3(3),
309–327. https://doi.org/10.1037/1082-989X.3.3.309
Martin, K. A., & Hall, C. R. (1995). Using mental imagery to enhance intrinsic motivation. Journal of
Sport and Exercise Psychology,17(1), 54–69. https://doi.org/10.1123/jsep.17.1.54
Martin, K. A., Moritz, S. E., & Hall, C. R. (1999). Imagery use in sport: A literature review and applied
model. The Sport Psychologist,13(3), 245–268. https://doi.org/10.1123/tsp.13.3.245
May, K., & Hittner, J. B. (2010). Reliability and validity of gain scores considered graphically. Perceptual
and Motor Skills,111(2), 399–406. https://doi.org/10.2466/03.PMS.111.5.399-406
Mellalieu, S. D., Hanton, S., & Thomas, O. (2009). The effects of a motivational general-arousal imagery
intervention upon preperformance symptoms in male rugby union players. Psychology of Sport
and Exercise,10(1), 175–185. https://doi.org/10.1016/j.psychsport.2008.07.003
Moher, D., Dulberg, C. S., & Wells, G. A. (1994). Statistical power, sample size, and their reporting in
randomized controlled trials. JAMA: The Journal of the American Medical Association,272(2),
122–124. https://doi.org/10.1001/jama.1994.03520020048013
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for systematic
reviews and meta-analyses: The PRISMA statement. PLoS Medicine,6(7). Article e1000097.
https://doi.org/10.1371/journal.pmed.1000097
Morris, T., Spittle, M., & Watt, A. P. (2005). Imagery in sport. Human Kinetics.
Munroe-Chandler, K. J., Hall, C. R., Fishburne, G. J., Murphy, L., & Hall, N. D. (2012). Effects of a cognitive
specific imagery intervention on the soccer skill performance of young athletes: Age group
20 B. A. SIMONSMEIER ET AL.
comparisons. Psychology of Sport and Exercise,13(3), 324–331. https://doi.org/10.1016/j.
psychsport.2011.12.006
Munroe-Chandler, K. J., Hall, C. R., Fishburne, G. J., & Shannon, V. (2005). Using cognitive general
imagery to improve soccer strategies. European Journal of Sport Science,5(1), 41–49. https://doi.
org/10.1080/17461390500076592
Murphy, S. M., & Martin, K. A. (2002). The use of imagery in sport. In T. S. Horn (Ed.), Advances in sport
psychology (2nd ed., pp. 405–439). Human Kinetics.
Nissen, J. M., Talbot, R. M., Thompson, A. N., & Van Dusen, B. (2018). A comparison of normalized gain
and Cohen’s d for analyzinggains on concept inventories. Physical Review Physics Education
Research,14(1). Article 010115. https://doi.org/10.1103/PhysRevPhysEducRes.14.010115
O, J., & Munroe-Chandler, K. J. (2008). The effects of image speed on the performance of a soccer task.
The Sport Psychologist,22(1), 1–17. https://doi.org/10.1123/tsp.22.1.1.
Paivio, A. (1985). Cognitive and motivational functions of imagery in human performance. Canadian
Journal of Applied Sport Sciences,10(4), 22S–28S.
Polanin, J. R., Tanner-Smith, E. E., & Hennessy, E. A. (2016). Estimating the difference between pub-
lished and unpublished effect sizes: A meta-review. Review of Educational Research,86(1), 207–
236. https://doi.org/10.3102/0034654315582067
Ramsey, R., Cumming, J., Edwards, M. G., Williams, S. E., & Brunning, C. (2010). Examining the emotion
aspect of PETTLEP-based imagery with penalty taking in soccer. Journal of Sport Behavior,33(3),
295–314.
Raudenbush, S. W. (2009). Analyzing effect sizes: Random-effects models. In H. Cooper, L. V. Hedges,
& J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (pp. 295–315). Russell
Sage Foundation.
R Core Team. (2008). R: A language and environment for statistical computing.http://www.R-project.
org/
Robin, N., Dominique, L., Toussaint, L., Blandin, Y., Guillot, A., & Le Her, M. (2007). Effects of motor
imagery training on service return accuracy in tennis: The role of imagery ability. International
Journal of Sport and Exercise Psychology,5(2), 175–186. https://doi.org/10.1080/1612197X.2007.
9671818
Rothstein, H. R., & Bushman, B. J. (2012). Publication bias in psychological science: Comment on
Ferguson and Brannick. Psychological Methods,17(1), 129–136. https://doi.org/10.1037/a0027128
Rothstein, H. R., Sutton, A. J., & Borenstein, M. (2005). Publication bias in meta-analysis: Prevention,
assessment and adjustments. John Wiley.
Schmidt, F. L., & Hunter, J. E. (2015). Methods of meta-analysis: Correcting error and bias in research
findings (3rd ed.). Sage.
Schuster, C., Hilfiker, R., Amft, O., Scheidhauer, A., Andrews, B., Butler, J., Kischka, U., & Ettlin, T. (2011).
Best practice for motor imagery: A systematic literature review on motor imagery training
elements in five different disciplines. BMC Medicine,9(75), 1–35. https://doi.org/10.1186/1741-
7015-9-75
Schweizer, G., & Furley, P. (2016). Reproducible research in sport and exercise psychology: The role of
sample sizes. Psychology of Sport and Exercise,23, 114–122. https://doi.org/10.1016/j.psychsport.
2015.11.005
Shackell, E. M., & Standing, L. G. (2007). Mind over matter: Mental training increases physical strength.
North American Journal of Psychology,9(1), 189–200.
Simonsmeier, B. A., Frank, C., Gubelmann, H., & Schneider, M. (2018). The effects of motor imagery
training on performance and mental representation of 7- to 15-year-old gymnasts of different
levels of expertise. Sport, Exercise, and Performance Psychology,https://doi.org/10.1037/
spy0000117
Spittle, M., & Kremer, P. (2010). Mental practice and the retention of motor learning: A pilot study.
Perceptual and Motor Skills,110(3), 888–896. https://doi.org/10.2466/PMS.110.3.888-896
Tanner-Smith, E. E., & Tipton, E. (2014). Robust variance estimation with dependent effect sizes:
Practical considerations and a software tutorial in Stata and SPSS. Research Synthesis Methods,5
(1), 13–30. https://doi.org/10.1002/jrsm.1091
INTERNATIONAL REVIEW OF SPORT AND EXERCISE PSYCHOLOGY 21
Tanner-Smith, E. E., Tipton, E., & Polanin, J. R. (2016). Handling complex meta-analytic data structures
using robust variance estimates: A tutorial in R. Journal of Developmental and Life-Course
Criminology,2(1), 85–112. https://doi.org/10.1007/s40865-016-0026-5
Toth, A. J., McNeill, E., Hayes, K., Moran, A. P., & Campbell, M. (2020). Does mental practice still
enhance performance? A 24 year follow-up and meta-analytic replication and extension.
Psychology of Sport and Exercise,https://doi.org/10.1016/j.psychsport.2020.101672
Van den Noortgate, W., López-López, J. A., Marín-Martínez, F., & Sánchez-Meca, J. (2013). Three-level
meta-analysis of dependent effect sizes. Behavior Research Methods,45(2), 576–594. https://doi.
org/10.3758/s13428-012-0261-6
Viechtbauer, W., & Cheung, M. W.-L. (2010). Outlier and influence diagnostics for meta-analysis.
Research Synthesis Methods,1(2), 112–125. https://doi.org/10.1002/jrsm.11
Wakefield, C., & Smith, D. (2012). Perfecting practice: Applying the PETTLEP model of motor imagery.
Journal of Sport Psychology in Action,3(1), 1–11. https://doi.org/10.1080/21520704.2011.639853
Weinberg, R., Jackson, A., & Seaboune, T. (1985). The effects of specific vs nonspecific mental prep-
aration strategies on strength and endurance performance. Journal of Sport Behavior,8(4), 175–
180.
Williams, S. E., & Cumming, J. (2011). Measuring athlete imagery ability: The sport imagery ability
Questionnaire. Journal of Sport and Exercise Psychology,33(3), 416–440. https://doi.org/10.1123/
jsep.33.3.416
Williams, R. H., & Zimmerman, D. W. (1996). Are simple gain scores obsolete? Applied Psychological
Measurement,20(1), 59–69. https://doi.org/10.1177/014662169602000106
Wright, D. J., McCormick, S. A., Birks, S., Loporto, M., & Holmes, P. S. (2015). Action observation and
imagery training improve the ease with which athletes can generate imagery. Journal of
Applied Sport Psychology,27(2), 156–170. https://doi.org/10.1080/10413200.2014.968294
Wright, C., & Smith, D. (2009). The effect of PETTLEP imagery on strength performance. International
Journal of Sport and Exercise Psychology,7(1), 18–31. https://doi.org/10.1080/1612197X.2009.
9671890
Zimmermann, D. W., & Williams, R. H. (1998). Reliability of gain scores under realistic assumptions
about properties of pre-test and post-test scores. British Journal of Mathematical and Statistical
Psychology,51(2), 343–351. doi:10.1111/j.2044-8317.1998.tb00685.x
22 B. A. SIMONSMEIER ET AL.
