Skip to main content

Models accounting for intention-behavior discordance in the physical activity domain: a user’s guide, content overview, and review of current evidence


There is a growing concern among researchers with the limited effectiveness and yet subsequent stagnation of theories applied to physical activity (PA). One of the most highlighted areas of concern is the established gap between intention and PA, yet the considerable use of models that assume intention is the proximal antecedent of PA. The objective of this review was to: 1) provide a guide and thematic analysis of the available models that include constructs that address intention-behavior discordance and 2) highlight the evidence for these structures in the PA domain. A literature search was conducted among 13 major databases to locate relevant models and PA studies published before August 2014. Sixteen models were identified and nine overall themes for post-intentional constructs were created. Of the 16 models, eight were applied to 36 PA studies. Early evidence supported maintenance self-efficacy, behavioral regulation strategies, affective judgments, perceived control/opportunity, habit, and extraversion as reliable predictors of post-intention PA. Several intention-behavior discordance models exist within the literature, but are not used frequently. Further efforts are needed to test these models, preferably with experimental designs.

One is hard-pressed to overstate the health benefits of regular moderate to vigorous intensity physical activity (PA). Benefits include a considerably reduced risk of most major chronic diseases such as heart disease, type 2 diabetes, several cancers and musculoskeletal disorders [1,2] as well as the prevention and rehabilitation of psychological disorders such as depression and anxiety [3,4]. Unfortunately, very few people in most developed nations engage in enough PA to reap a considerable effect on their health. For example, recent population-level assessments of PA using accelerometry suggest that over 80% of North Americans do not meet the guidelines for public health [5,6]. Clearly, PA promotion efforts are needed.

Theoretical understanding of the determinants behind PA has been a line of research inquiry for over 30 years [7]. The premise behind this research is that a sound understanding of PA determinants will aid in intervention success and theories represent an organizing framework to provide structure, function and common nomenclature to critical variables under study [8,9]. Many of the most researched theories applied to understanding PA include intention as the most proximal antecedent to PA. These include, but are not limited to, the theory of planned behavior [10], theory of reasoned action [11], protection motivation theory [12], social cognitive theory [13], variants of self-determination theory [14] and even stage models such as the transtheoretical model, where stage of change is an intention-behavior hybrid construct [15].

Tests of the intention construct and its relationship with PA have solid validation. For example, the most recent meta-analysis of the theory of planned behavior applied to PA showed r = .48 [16], which places this relationship within the medium-sized effect range [17]. Still, the finding also suggests that 77% of the variance in PA is unexplained. The relationship is also further attenuated when examining change in PA (i.e., controlling for past PA), which is arguably far more accurate when attempting to understand intention and its role in behavior change [8,18]; the relationship between PA and intention reduces to r = .22, which suggests a borderline meaningful effect [17].

Examinations of the absolute, rather than relative, value of intention-behavior relations have also shown considerable discordance. For example, experimental manipulations that increase PA intention (d = .45) result in much lower, and clinically less meaningful increases in PA (d = .15) [19]. Dichotomization of the intention and PA relationship around public health guidelines also showed that 48% of intenders failed to follow-through with PA [20]. Perhaps most important, is the lowered practical value of theories that place intention as the proximal antecedent of PA. It is extremely common for participants in PA interventions to report to the trial with high intentions at baseline (i.e., often the driving reason for study participation), yet low PA. This phenomenon poses a challenge to intention-based theories because those intentions are considered the proximal variable to behavioral enactment. Taken together, this line of research has prompted researchers to call intention-behavior discordance “the intention-behavior gap.”

The point here is not to doom intention-based theories or the intention construct. Indeed, research into intention-PA discordance clearly shows that intention is a necessary construct for behavioral enactment; almost no one reports performing PA with low intention [20]. Still, almost half of intenders do not follow-through with their intentions, which suggests that intention is a necessary, but insufficient construct for many when enacting behavior. An understanding of the translation of intentions into behavior therefore seems paramount to improve upon our theoretical frameworks in PA [21].

The separation of intention formation from intention translation, or what is sometimes referred to as action control [22], is not new. Indeed, conceptualizations of the distinctions between these concepts can be traced to the very early 20th century [23] and they were contemplated carefully in the mid-20th century [24], but models of intention formation have received the bulk of attention in the research community since that time. Contemporary research on the intention-behavior “gap,” however, has begun to shift some researchers and theorists back to an investigation of post-intentional constructs. In this relatively early phase of re-interest into intention-translation, it would seem helpful to collect the models available to PA researchers and practitioners that attempt to account for intention-behavior discordance. Comprehensive reviews of theories related to PA have mentioned some of these intention-translation models in passing commentary e.g., [25,26] but we are unaware of a collective review of these models. Thus, the purpose of this review was to 1) provide a user’s guide to the available models, frameworks, and theories1 that specifically include a post-intentional construct or constructs with a content analysis of the various constructs proposed to account for intention-behavior discordance and 2) highlight the available evidence for these structures when applied to the PA domain. The review is meant to be more descriptive than evaluative of these models at this phase and ultimately to raise awareness and empower PA researchers and practitioners with options for models that attempt to account for intention-behavior discordance.


Eligibility criteria

Our literature search unfolded in a two-step process based on the purposes of the paper. First, we needed to identify what models offer a pre- and post-intentional conceptualization. To achieve this aim, an appropriate model for this review had to include the following criteria: First, the model had to be identified by a title or nomenclature to its identity. Thus, models tested as mere one-off tests of intention-based models (e.g., does personality mediate intention) were excluded. Second, the model had to include the intention construct within its frame in order to specifically demonstrate a post-intentional mechanism and intention needed to be a cognitive construct. Thus, models where intention formation may be implied, but not specifically articulated such as control theory [27] or dynamic action theory [28] were excluded. Further, models like the transtheoretical model [15], where the delineation between intent, plans and behavior are not clear (e.g., the preparation stage is marked by intent but also by some behavioral action and organizational/preparatory behaviors) were also excluded. Third, the model needed to have at least one construct where the central mechanisms were theorized as post-intentional. This could take three forms: 1) constructs proposed as mediators of intention-behavior relations, 2) constructs proposed as moderators of intention-behavior relations, and 3) constructs unrelated to intention in the theory that directly determine behavior. The latter consideration above was the aspect that needed the most careful scrutiny. Constructs in models that are not specifically stated as independent from intention formation did not meet our inclusion criteria, such as theory of planned behavior’s perceived behavioral control construct, where perceived behavioral control is hypothesized as an antecedent to intention and a potential direct determinant to behavior [10]. The second step in our review was to identify the application of all models found in step one to the PA domain. Eligible studies included the following: 1) the use of the identified intention-based model with post-intentional constructs and 2) a measurement of PA as a primary outcome. Studies were delimited in both searches to published works written (or translated) to English.

Search strategy

A comprehensive literature search was conducted to identify potential models that integrated post-intentional constructs and have been previously applied to health behaviors. Health behaviors of interest included PA/exercise, healthy eating, smoking cessation, flossing, sunscreen use, vaccination, HIV prevention, or cancer screening. Major databases (Academic Search Complete, Academic Search Premier, AgeLine, CINAHL, Health Source, Medline, Alt HealthWatch, Health Technology Assessment, PsychINFO, PubMed, Social Sciences, SPORTDiscus and Web of Science) were used to search for any articles published prior to August 2014. A combination of the following key words that were used included: model, theory, framework, health behavior, physical activity, exercise, dieting, nutrition, healthy eating, smoking, flossing, cancer screening, HIV prevention, condom use, vaccination, immunization, volition, action, behavior, intention, post-intention, intention-behavior gap, mediator, and moderator. One author conducted the search, and reference lists of included studies were manually cross-referenced. Previously identified models from the authors’ personal records were added to the search. To enhance the comprehensiveness of our review, we also examined all 83 theories outlined in Michie and colleagues’ book, “ABC of Behavior Change Theories” [9]. This book has amalgamated many theories published to date and authored by experts in various fields.

The secondary search to identify the applications of the identified models in PA utilized the aforementioned databases. Keywords used in this search included the specific name of the model (e.g., health action process approach), exercise, and physical activity. All the reference lists of the studies included were carefully inspected to locate any additional studies.

Data abstraction and analysis

Each author evaluated the identified models separately to determine its eligibility for the review. Once deemed eligible and a consensus was reached, all constructs relevant to intention-behavior discordance were themed. Our thematic analysis approach followed an iterative process. We first constructed a list and definition of the properties of prototypical constructs present in the most popular models in health and social behavior to use as a frame of reference [13,29,30]. Constructs from the identified models were then compared to this list and subsequently scored as one of these constructs or added to the list as an additional construct. Once an initial set of themes had been created, analysis shifted toward identifying broader patterns– that is, establishing the number of themes deemed sufficient to capture the essence of the main constructs in each model, followed by creating a label and description of each. Through this process, themes deemed sufficiently similar to one another were combined. All constructs were abstracted and categorized into themes independently and then compared to reach consensus.

In our subsequent analysis, the abstracted information included authors, country, sample (number, age, and gender), study design, model/framework/theory used, the use of the full model and post-intentional constructs, and any significant post-intentional predictors. Evidence was organized in observational and experimental categories and significant and meaningful associations by construct were tallied. Themes were created if three or more studies were present based on the prior review methodology by Rhodes and colleagues e.g., [31-33]. Based on Sallis et al.’s [34] rubric for determining an association among studies, a construct was considered to have a positive association (+) if greater than 59% of studies supported prediction of PA (or PA change); a negative association (−) if greater than 59% of studies supported a decrease in PA; inconclusive (?) if 34-59% of studies found any association, and no association (0) if less than 34% of studies showed any association. When analyzing the variables, both 1) statistical significance (p < .05) and a meaningful effect size [35,36] (d > .19; r > .09; OR > 1.49) needed to be present in order for a predictor to have either a positive or negative association with PA. Studies where effect size could not be determined were scored on significance value alone.


Details from the literature search for intention discordance models are described in a flow-diagram in Figure 1. The initial search yielded 3,388 potentially relevant records. Most papers were excluded because the study contained no indication of a specific model, or contained a model that did not address intention-behavior discordance. Twenty-six abstracts and full-text reports were obtained and reviewed. A total of 16 sources passed the eligibility criteria and were included in the initial part of the review.

Figure 1
figure 1

Initial search for models, frameworks, and theories that address intention-behavior discordance.

All of these models were published in English between 1980 and 2014. The models originated from a range of countries, which included Germany, United States, Canada, Australia, Netherlands, Denmark and the United Kingdom. Most of the theoretical models, namely the ecological model for preventing type 2 diabetes in minority youth [37], health action process approach (HAPA) [38], precaution adoption process model [39], and temporal self-regulation theory [40], have simultaneously focused their initial efforts on a variety of health behaviors (e.g., diet and nutrition, PA, smoking cessation). Whereas, the integrated behavior change model [41], motivation-volition (MoVo) process model [42], and the action control framework, now referred to as the multi-process action control (M-PAC) framework [43], have been solely focused on PA behavior. Other models such as the integrated change (I-Change) model [44], information-motivation-behavioral skills model [45] and plans, responses, impulses, motives, and evaluations (PRIME) theory [46] have targeted health communication, HIV prevention and smoking cessation respectively. The action theory model of consumption [47] and motivation-ability-opportunity-behavior model [48] were originally conceptualized with consumerism as the targeted behavior. Finally, theories like the action control theory [22], Rubicon model of action phases [49], volitional model of goal-directed behaviors [50], and theory of interpersonal behavior [51] were generated as generic behavior models.

Content analysis of intention-behavior discordance constructs

The two reviewers identified nine overall themes through content analysis of each theory (see Table 1). Each theme is detailed below:

Table 1 Identified theories, model, and frameworks and intention-behavior discordance constructs (n = 16)

Volitional regulation behaviors

Behaviors used to maintain or hone intentions featured prominently in 11 of the models identified [22,38,41-47,49,50]. These are described either broadly, with examples such as self-monitoring, scheduling, enlisting support, prioritizing and problem-solving around other behaviors as strategies an individual may use to maintain intentions [22,42,43,45,49,50] and/or specifically in terms of action plans or implementations intentions [38,41,44] and coping plans used to problem solve around difficulties in enacting the behavior [38].


Affect-based constructs were present in four of the reviewed models [22,43,46,50]. Affective judgments (e.g., enjoyment, pleasure) about the behavior itself and the motivational influence these have to maintain or abandon intentioned behavior were featured in action control theory [22], the volitional model of goal-directed behaviors [50], M-PAC [43], and PRIME theory [46]. Affect control in the form of shunting undesired general affective states in order to maintain one’s intentions featured in action control theory [22].

Self-efficacy/behavioral skills

Different types of self-efficacy variants are present in eight of the models reviewed [37-39,43-45,47,48]. The knowledge-attitude-behavior model [37], motivation-ability-opportunity-behavior model [48], and the precaution adoption process model [39] suggest that self-efficacy is a moderator of intention translation, with high self-efficacy needed to translate intentions into action. M-PAC suggests that control in the form of skills/ability, akin to task self-efficacy [52], impacts intention-translation with a threshold effect, where higher abilities may be needed to translate an intention into behavior than form an intention [43]. A similar construct of skills was proposed to mediate and moderate the intention and behavior relationship in the I-Change model [44] and mediate motivation-behavior relations in the information-motivation-behavioral skills model [45]. By contrast, the HAPA [38] suggests that two types of self-efficacy aid in translating intentions into behavior. Maintenance self-efficacy represents the confidence one can perform a behavior given various barriers. Recovery self-efficacy represents an individual’s confidence in performing the behavior under relapse or brief periods of non-performance. Finally, the theory of consumption positions perceived control as an antecedent of desire to achieve an intention, implementation intentions, volitional control strategies used in trying the behavior, and behavior itself [47].

Outcome expectations

Action control theory [22], PRIME theory [46], and precaution adoption process model [39] identified the use of outcome expectations in the role of translating intentions into behavior. These models suggest that conscious reminders of the expected outcomes from a behavioral act can be used to help bolster motivation.

Selective attention

Two of the models identify selective attention processes in maintaining intentions and following thorough with behavior [22,42]. This concept features prominently in action control theory [22] where selective processing of new information and competing intentions, and inhibition from over-processing information about the intended behavior are hypothesized as critical in order to translate intentions into behavior. Gӧhner et al.’s [42] MoVo process model also considers selective processing as crucial to moderating intention-behavior relations in the form of intention-shielding properties.

Endogenous factors

Stable individual differences featured in five of the models of intention-behavior discordance [22,40,43,46,51]. The theory of interpersonal behavior [51] considers individual arousal level as an independent system from intention that may facilitate or impede behavior. Action control theory [22] suggests that individuals have predispositions toward being either action-oriented and following through on intentions, or state oriented and maintaining the status quo. M-PAC [43] suggests that PA intention translation is affected by the personality traits of conscientiousness and extraversion, where those higher in these personality traits are more likely to follow-through with their intentions due to dispositional drive systems of achievement/order and energy-level respectively. Another unique consideration is Hall and Fong’s [40] construct of self-regulatory capacity in their temporal self-regulation theory, a biologically-based system of physiological energy and executive function capacity, that is proposed to moderate intentions and exert its own independent effect on behavior. Finally, West’s PRIME theory [46] suggest that dispositions are characteristics of the motivational system that govern its operation in stability and context sensitivity, thus affecting how intention gets translated into action.

External factors

Environmental and other facilitating/inhibiting conditions feature in eight of the models reviewed [39,40,43,44,46,48,50,51]. In Triandis’ [51] theory of interpersonal behavior and Baggozi’s [50] volitional model of goal-directed behaviors, facilitating (or inhibiting) environmental conditions are proposed to determine behavior as an independent system from intention. The M-PAC model suggests that opportunity to act (i.e., time and access) impacts intention-translation with a threshold effect, where greater opportunities to act are needed to translate an intention into behavior than form the original intention [43]. The precaution adoption process model [39] highlights that time, resources, competing opportunities all impact translation of intention into behavior. Finally, de Vries et al. [44] in their I-Change model, Öllander and Thørgersen [48] in their motivation-ability-opportunity-behavior model, West’s [46] PRIME theory, and Hall and Fong’s [40] temporal self-regulation theory suggest that external barriers (e.g., environmental and social) will moderate the intention-behavior relationship, with more barriers resulting in an attenuation of the intention-behavior link.


Six models suggest processes below conscious thinking are critical [40,43,47,48,51,53]. The theory of interpersonal behavior [51] proposes that habit, defined as automatic responses to cues from a patterned and learned behavior will determine behavior independent of intention. Bagozzi’s [47] model of consumption suggests that frequency and recency of behavior impact both the behavioral processes employed to enact a behavior and the behavior itself. The M-PAC [43] and motivation-ability-opportunity-behavior model [48] suggest that habit helps bind intended behavior to behavioral action. PRIME theory [46] highlights the importance of routine to the translation of plans (i.e., intentions) into action. Finally, Hall and Fong’s [40] temporal self-regulation theory proposes that behavioral pre-potency, a construct almost identical in conception to habit, will moderate the intention-behavior relationship and exert its own influence on behavior.


Four models suggest that role identity formation with a behavior is related to intention translation [43,47,50,53]. The M-PAC model [43] and Bagozzi’s volitional model of goal-directed behaviors [50] and theory of consumption [47] suggest that identity serves to preserve an intended behavior through the process of reducing affective/cognitive dissonance in contexts that trigger an awareness of one’s role identity (i.e. exerciser) with a discrepant action (e.g., not exercising). PRIME theory [46] contends that identity serves in a similar capacity for translating plans into action. Specifically, identity is thought to be the ultimate source of self-regulation.

Evaluation of the models of intention-behavior discordance applied to physical activity

For the subsequent analysis, the literature search generated 181 relevant studies. Studies were predominantly excluded due to no concentration on PA behavior, no relevancy to intention-behavior discordance, or published in another language other than English. Overall, a total of 41 full-text articles were reviewed and assessed for eligibility. Of these, 36 studies were eligible for the analysis. Further details are outlined in Figure 2.

Figure 2
figure 2

Search for individual studies that utilize models, frameworks with post-intentional constructs.

Only eight of the 16 previously identified theories were applied to PA (see Table 2 and Additional file 1: Table S1). Most commonly used, the HAPA model was applied to 12 observational studies and three intervention-based studies, while the M-PAC framework was used in 11 correlational studies and one experimental study. Less commonly applied theoretical models were the theory of interpersonal behavior [54], precaution adoption process model [55,56], I-Change model [57], information-motivation-behavioral skills model [58,59], MoVo process model [60,61], and temporal self-regulation theory [62]. Of these studies, five were observational and four were experimental. Due to the limited applications of these models, only the M-PAC model, and HAPA model will be discussed in further detail.

Table 2 Descriptive summary of the individual studies included in the analysis (n = 36)

Health action process approach

Schwarzer’s [38] HAPA model was developed with specific aims to address intention-behavior discordance among health behaviors. The model differentiates between a pre-intentional phase of motivation and a post-intentional phase of volition for behavioral enactment. Action self-efficacy (perceived capability to perform the behavior), perception of risk from inaction, and outcome expectancies (expected consequences) are considered the determinants of intention, while coping-self-efficacy (confidence to handle barriers), action planning (detailed instructions of the behavioral enactment), coping planning (plans to handle barriers), and recovery self-efficacy (confidence to handle resume the behavior after set-backs) are considered the post-intentional determinants of behavior. The HAPA model also suggests a process of initiation to maintenance, where recovery self-efficacy is critical during behavioral maintenance.

Table 3 details the findings by HAPA constructs of the 12 studies, yet 14 independent samples that have formally applied the model to predict PA [63-74]. Three studies have tested the entire original model suggested by Schwarzer [38] in a single study i.e., [67,71,74], and enough studies are available to evaluate preliminary findings for the utility of each construct.

Table 3 Summary of the post-intentional predictors in physical activity

Eight of nine studies have supported the predictive path of action self-efficacy on intention [63-70,74], most with large effect sizes (i.e., β > .50), demonstrating this as a reliable finding. Nine [63,64,67-69,71,73,74] and two [65,68] of 13 samples have shown support for the path of outcome expectations and risk perceptions on intention, respectively. The results suggest that outcome expectations may be useful in understanding PA intention, while risk perceptions may not.

Of the key constructs accounting for intention-behavior discordance, maintenance self-efficacy has had relatively strong support with eight [63,65,66,68,69,73,74] significant tests of 12 possible samples [64,67,71,72] (see Table 3). By contrast, recovery self-efficacy has shown a significant path in two [67,72] of six samples [70 samples 1 and 2, 71, 74]. Action planning was measured in eight samples [63,64,66,67,69,70,73], yet only two of these showed a significant path [66,73]. Only three samples have specifically applied coping planning and two supported a significant path to PA [70 samples 1 and 2] while the other was not significant or of meaningful magnitude (β < .10) [67]. Finally, of the five samples that employed an aggregate of coping and action planning [65,68,71,74], only one sample indicated this as a significant predictor of PA [71]. The results support maintenance self-efficacy as a post-intentional predictor of PA and suggest that coping planning may have utility, but action planning and an aggregate of action planning and coping planning does not appear to be useful to predict for intention-behavior discordance as proposed by the HAPA model.

Three intervention studies have applied the HAPA to promote PA [75-77]. All three have intervened upon the action and coping planning constructs and not the maintenance and recovery self-efficacy constructs. These studies have varied in participants from pregnant women [77] to obese patients [75] and a general population sample [76]. Two of the three studies have shown both statistically significant and meaningful (d > .20) increases in PA compared to control groups [76,77], yet only Gaston and Prapravessis [77] were able to demonstrate that the changes in PA were from changes in planning. The study by Ströbl and colleagues did not show meaningful effect-sized changes in PA, yet the difference in this study may be attributable to the six and 12 month post-intervention assessment compared to both Lippke et al. [76] and Gaston and Prapravessis [77], who examined planning effectiveness on PA over a one month period. Taken together, the results support HAPA’s potential effectiveness of action and coping planning, at least over a short duration, in order to increase PA.

Multi-process action control model

The M-PAC conceptual model proposed around the action control framework by Rhodes and de Bruijn [43] defines intention as a binary decisional choice variable and not the intention strength (and breadth of the motivational spectrum) definition taken in many intention-based theories [78]. Intention choice is formed by instrumental attitude (or outcome expectations), affective attitude (or experiential expectations) and perceived control constructs of ability (physical skill/movement without injury) and opportunity (time/access). Subsequent translation of intention choice into PA, called action control, is thought to be determined by regulatory behaviors (e.g., coping planning, enlisting support, self-monitoring), particularly during the initial adoption of the behavior. Continuation of the behavior is thought to include the addition of more reflexive means of action control via habit (responses to cues) and identity (responses due to role activation) development as one maintains a behavioral pattern. Furthermore, affective attitude and perceived control constructs are considered non-linear during the intention-formation to action control process, where much higher values are considered necessary for successful action control than mere intention formation. Personality traits of conscientiousness and extraversion are considered potential endogenous antecedents of action control, where those higher in these personality traits are more likely to follow-through with their intentions due to dispositional drive systems of achievement/order and energy-level respectively. Environmental and social context are considered subsumed (i.e., mediated) by affective attitude and perceived control constructs during the action control process which may facilitate or inhibit behavior.

Table 3 details the aggregate results for 11 prior prediction-based tests of the action control framework [79-89] and one experimental test [90]. No single study to date, however, has examined the entire conceptual model suggested by Rhodes and de Bruijn [43].

Tests have supported the nonlinear effects of affective attitude and perceived control in 7/8 and 8/10 tests respectively [see 43 for the full review]. Further, instrumental attitude failed to predict action control in nine of 10 samples, supporting its absence in the conceptual model after the initial formation of the intention. Self-regulatory constructs were assessed in five tests to evaluate action control. Four of the five studies showed significant prediction including the behavioral processes of change [86,87] (i.e., strategies such as rewards, self-monitoring, enlisting support, and creating stimulus control), regulation over other leisure behaviors [85] and coping planning [90] but not action planning [79]. Three studies employed a habit construct (i.e., enacting PA from external cues, starting PA without deliberation) to predict action control and found support for the construct [79,85,88]. Finally, three studies employed personality trait measures of extraversion (i.e., sociability, positive affect, assertiveness, preference for lively activity) and conscientiousness (i.e., industriousness, orderliness, self-discipline) to predict action control framework and found two in support of extraversion [84,89] but only one test in support of conscientiousness [80]. Results examining the identity construct and the difference among constructs between adoption and maintenance have not been evaluated sufficiently to produce an outcome theme at present.


There is a growing concern among theoretical researchers with the limited effectiveness and yet subsequent stagnation of theories applied to health behaviors like PA [21,91-95]. At the top of this list are theories that position intention as the proximal predictor of behavior due to the well-established intention-behavior gap [19,20,78,96]. In the spirit of trying to move forward in theory development and application, the purpose of this review was to 1) provide a user’s guide and thematic analysis to the available models that specifically include a construct to explain intention-behavior discordance and 2) highlight the available evidence for these structures when applied to the PA domain.

Our review yielded 16 models that have positioned constructs in pre-intentional and post-intentional positions. When taken into context of the gamut of possible health behavior theories [9], this represents approximately 19% of known theoretical frameworks. Thus, a sizeable proportion of models are available for researchers attempting to account for the intention-behavior gap. Many of these models are relatively new – eight have been proposed within the last 10 years. Nevertheless, some have been available since the 1980s [22,39,49,51] or early 1990s [45,48,50]. Researchers have had the opportunity to test these theories in the PA domain for quite some time.

Despite 16 different models for researchers to choose from, our content analysis of constructs suggested considerable overlap in the kinds of factors that are being proposed to account for intention-behavior discordance and some redundancies with general intention models. The hallmark of many post-intentional models is the inclusion of volitional regulatory behaviors used to maintain or hone intentions. Theorists suggest that people who form intentions, need to then become strategic in order to implement their intentions across the backdrop of competing forces for their attention, motivation, and time. Eleven of these models included these behavioral regulation constructs, yet they differed from very specific concepts such as action plans or implementations intentions to a more general array of behaviors such as self-monitoring, scheduling, enlisting support, prioritizing and problem-solving. Volitional behaviors used to maintain or hone intentions have sound empirical evidence in the PA domain generally [97-101] and within a stage-based model such as the transtheoretical model of behavior change [102]. They also have high face validity, as participants who report for PA intervention trials often already have high intentions and seek strategic advice for the translation of these intentions into behavior [43].

Two post-intentional elements that were suggested in nearly half of the models included external conditions and/or endogenous factors. While these are at opposite ends of the spectrum in terms of individual vs. environment, they both share more immutable qualities that may either facilitate or thwart intention translation. For example, external conditions represent the opportunities an individual may have to enact PA as represented by the social and physical environment. This concept takes social ecology into consideration within post-intentional models, where its omission in several intention-based models has often been a criticism [103]. The concept also has support in the PA domain outside of these models, where a factor such as proximity to recreation facilities has been shown to moderate intention-behavior relations [104-106]. Endogenous factors represent individual differences in organizational/executive function, cravings and energy/arousal that may thwart or facilitate intention translation or enactment of behavior. Endogenous factors as constructs used to explain health behavior are not unique to post-intentional models, but the operation of these constructs take neuroscience, broadly construed, into consideration as a factor explaining the intention-behavior gap. Disposition has been shown to moderate the PA intention-behavior gap in several tests outside of these theories see [107] for a review, but application tests of several of the constructs proposed in these theories are needed.

More reflexive constructs such as habit (responses to cues) and identity (responses from role activation) were featured in about a third of the models. These are interesting concepts, not traditionally employed in many popular intention theories. These constructs suggest that intention translation may be tied to less reasoned or pre-meditated processes [108]. Both have had some evidence in the PA domain outside of these post-intentional models [109,110] but require sustained research attention.

Finally, specific motivational concepts such as specific types of self-efficacy, outcome expectations and affective judgments were also identified in many of the models. These do not represent particularly new concepts from prior intention-based models but the implication for these types of constructs is that intention translation requires either different forms of motivation than intention formation or a higher threshold of these motivational constructs. The presence/absence of these factors underscored some of the differences among these models in terms of how intention was operationalized. For example, some theories considered intention as the definitive motivational variable (intention strength), akin to the definition put forth by Fishbein and Ajzen [11]; thus, post-intentional constructs needed to represent concepts outside the motivational domain e.g., [40,42,49]. By contrast, other theorists positioned intention more within the dictionary meaning of an aim toward a behavior (intention choice) [111], thus allowing for motivational variables to affect that aim during intention translation [22,43,46,47,50]. Applied researchers will need to carefully consider how they view intention in order to select a theoretical frame.

The definition of behavior, the application of the theory for intervention, and the specific mechanisms accounting for intention-behavior discordance were also somewhat different across the various models. For example, HAPA [38], temporal self-regulation theory [40], and M-PAC [43] give some explicit consideration of adoption vs. maintenance behavior and how different constructs may affect behavior across time. Temporal self-regulation theory [40] appears to be an explanative model of behavior first and foremost, while information-motivation-behavioral skills model [45] is a simplified model for interventionists. Most of the other models were positioned somewhere in between these approaches. Finally, some models position their post-intention constructs as moderators of intention-behavior relations e.g., [37,41,48], others position them as direct mediators e.g., [38], while some position these constructs as separate systems impacting behavior independent of intention e.g.,[51]. Indeed, our content analysis suggested that Temporal self-regulation theory [40] and the theory of interpersonal behavior [51] contained relatively similar post-intentional constructs overall, but their proposed mechanisms were markedly different. Thus, applied researchers need to give attention to the specific suggestions of how to model each theory.

Our second purpose of the paper was to review and highlight the available evidence for these 16 models when applied to the PA domain. Perhaps the most important finding from this second purpose was a demonstration of how little these models have been applied to understand PA. Our application of theory in PA tends to extend to only a handful of approaches [112], none of which are among these models. We had anticipated this outcome, and it formed one of the central reasons for writing the paper as a user’s guide.

Thirty-six studies were identified for the 16 models, but 27 of these were from two of the models (HAPA and M-PAC). Eleven of the models had either one or zero applications in the PA domain making any kind of assessment too preliminary. There was enough research available to evaluate the early evidence for both M-PAC [43] and HAPA [38] in the PA domain. M-PAC is more of a methodological framework than a theory at this juncture and the results herein were actually used to create the proposed conceptual model post-hoc, as opposed to validating the model. Future tests and the use of its complete set of constructs are definitely warranted. Thus far, there is evidence that affective judgments, perceived control/opportunity, habit, and extraversion can reliably predict PA intention translation but there is not enough evidence to evaluate the adoption to maintenance transition proposed in the conceptual model or the role of identity. There is also not enough experimental work with the model to warrant an appraisal. Unlike M-PAC, HAPA has had much better independent assessment outside of the original studies used to create the model, but application of the full model was also limited. Interestingly, nearly half of the samples employing HAPA included clinical populations, suggesting good heterogeneity of its application although no marked difference in findings was present between clinical and nonclinical samples at this time. Of the key constructs accounting for intention-behavior discordance, maintenance self-efficacy has had relatively strong support as a predictor PA independent of intention. This suggests that how people cope with PA barriers may be essential to successful behavioral enactment. Action planning, on the other hand, has shown generally weak to negligible effects. Coping planning has fared better as a mediator of intention than action planning but its application has been limited. Experimental work with the model has also supported HAPA’s planning constructs (at least action planning and coping planning in tandem) for short-term PA change, but no studies were present to evaluate the maintenance and recovery self-efficacy constructs.

Taken together, intention-based theories have dominated the PA research landscape, and there has been a sluggishness to embrace models that attempt to account for intention-behavior discordance. Clearly, there are several models that attempt to account for this discordance within the literature, but they have not been used and validated. The reasons for the dominance of a few select theories in PA research may be from tradition (researchers trained from a similar set of supervisors), convenience (easier to test and publish with models of high use), lack of awareness, or general lack of innovation. Whatever the case, the models identified in this review propose several variables such as volitional strategies, social and environmental conditions, disposition, habit, identity, as well as affect, PA skill and selective attention processes that are not as well-defined in our traditional intention theories. Whether these models will serve to improve our interventions and/or explain PA better than the current state of research is unknown, yet early work with HAPA has shown some promise. At a minimum, many of these models may serve to better integrate extant interventions that already focus on volitional strategies, environmental change, social processes/dynamics and individual differences, etc. [100]. The results of this paper suggest there are certainly models that warrant testing, preferably with experimental designs, in the future.

Footnote 1: To simplify, we refer to models, theories, approaches or frameworks as “models” generically from this point forward.


  1. 1.

    Warburton DER, Charlesworth S, Ivey A, Nettlefold L, Bredin SSD. A systematic review of the evidence for Canada's physical activity guidelines for adults. Int J Behav Nutr Phys Act. 2010;7:39.

    Google Scholar 

  2. 2.

    Warburton DER, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence. Can Med Assoc J. 2006;174:801–9.

    Google Scholar 

  3. 3.

    Asmundson GJG, Fetzner MG, DeBoer LB, Powers MB, Otto MW, Smits JAJ. Let's get physical: A contemporary review of the anxiolytic effects of exercise for anxiety and its disorders. Depress Anxiety. 2013;30:362–73.

    Google Scholar 

  4. 4.

    Mammen G, Faulkner G. Physical activity and the prevention of depression: a systematic review of prospective studies. Am J Prev Med. 2013;45:649–57.

    Google Scholar 

  5. 5.

    Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian health measures survey. Health reports. Statistics Can. 2011;22(21):21–8.

    Google Scholar 

  6. 6.

    Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40:181–8.

    Google Scholar 

  7. 7.

    Dishman RK. Exercise adherence: its impact on public health. Champaign, IL: Human Kinetics; 1988.

    Google Scholar 

  8. 8.

    Rhodes RE, Nigg CR. Advancing physical activity theory: A review and future directions. Exerc Sports Sci Rev. 2011;39:113–9.

    Google Scholar 

  9. 9.

    Michie S, West R, Campbell R, Brown J, Gainforth H. ABC of Theories of Behaviour Change; in press.

  10. 10.

    Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. 1991;50:179–211.

    Google Scholar 

  11. 11.

    Fishbein M, Ajzen I. Belief, Attitude, Intention, and Behavior. Don Mills, NY: Addison-Wesley; 1975.

    Google Scholar 

  12. 12.

    Rogers RW. Cognitive and physiological processes in fear appeals and attitude change: A revised theory of protection motivation. In: Cacioppo JT, Petty RE, editors. Social Psychophysiology. New York: Guilford Press; 1983. p. 153–76.

    Google Scholar 

  13. 13.

    Bandura A. Health promotion from the perspective of social cognitive theory. Psychol Health. 1998;13:623–49.

    Google Scholar 

  14. 14.

    Hagger M, Chatzisarantis NLD. Integrating the theory of planned behaviour and self-determination theory in health behaviour: A meta-analysis. Br J Health Psychol. 2009;14:275–302.

    Google Scholar 

  15. 15.

    Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Health Promot. 1997;12:38–48.

    CAS  Google Scholar 

  16. 16.

    McEachan RRC, Conner M, Taylor NJ, Lawton RJ. Prospective prediction of health-related behaviors with the theory of planned behavior: A meta-analysis. Health Psychol Rev. 2011;5:97–144.

    Google Scholar 

  17. 17.

    Ferguson CJ. An effect size primer: A guide for clinicians and researchers. Proffesional Psychol Res Practice. 2009;40:532–8.

    Google Scholar 

  18. 18.

    Weinstein ND. Misleading tests of health behavior theories. Ann Behav Med. 2007;33(1):1–10.

    Google Scholar 

  19. 19.

    Rhodes RE, Dickau L. Meta-analysis of experimental evidence for the intention-behavior relationship in the physical activity domain. Health Psychol. 2012;31(6):724–7.

    Google Scholar 

  20. 20.

    Rhodes RE, de Bruijn GJ. How big is the physical activity intention-behaviour Gap? a meta-analysis using the action control framework. Br J Health Psychol. 2013;18:296–309.

    Google Scholar 

  21. 21.

    Sniehotta FF, Presseau J, Araújo-Soares V. Time to retire the theory of planned behavior. Health Psychol Rev. 2014;8:1–7.

    Google Scholar 

  22. 22.

    Kuhl J. Motivational aspects of achievement motivation and learned helplessness: Towards a comprehensive theory of action control. In: Maher BA, Maher WB, editors. Progress in Experimental Personality Research. Volume 13. New York: Academic Press; 1984. p. 99–171.

    Google Scholar 

  23. 23.

    Ach N. Uber die Willenstatigkeit und das Denken. Gottingen: Vandenhoeck & Ruprecht; 1905.

    Google Scholar 

  24. 24.

    Lewin K. Intention, Will, and Need. In: Rapaport D, editor. Organization and Pathology of Human Thought. New York: Columbia University Press; 1951. p. 95–153.

    Google Scholar 

  25. 25.

    Armitage CJ, Conner M. Social cognition models and health behaviour: A structured review. Psychol Health. 2000;15:173–89.

    Google Scholar 

  26. 26.

    Biddle SJH, Fuchs R. Exercise psychology: A view from Europe. Psychol Sport Exercise. 2009;10:410–9.

    Google Scholar 

  27. 27.

    Carver CS, Scheier MF. Control theory: A useful conceptual framework for personality–social, clinical, and health psychology. Psychol Bull. 1982;92:111–35.

    CAS  Google Scholar 

  28. 28.

    Atkinson JW, Birch DA. A Dynamic Theory of Action. New York: Wiley; 1970.

    Google Scholar 

  29. 29.

    Fishbein M, Triandis HC, Kanfer FH, Becker M, Middlestadt SE, Eichler A. Factors influencing behavior and behavior change. In: Baum A, Revenson TA, editors. Handbook of Health Psychology. Mahwah, New Jersey: Lawrence Erlbaum Associates; 2001. p. 3–17.

    Google Scholar 

  30. 30.

    Rhodes RE, Mark R. Social Cognitive Theories. In: Acevedo EO, editor. The Oxford Handbook of Exercise Psychology. New York: Oxford University Press; 2012. p. 273–94.

    Google Scholar 

  31. 31.

    Gustafson S, Rhodes RE. Parental correlates of child and early adolescent physical activity: A review. Sports Med. 2006;36:79–97.

    Google Scholar 

  32. 32.

    Kirk M, Rhodes RE. Occupation correlates of adults’participation in leisure-time physical activity: A systematic review. Am J Prev Med. 2011;40:476–85.

    Google Scholar 

  33. 33.

    Rhodes RE, Temmel C, Mark R. Correlates of adult sedentary behaviour: A systematic review. Am J Prev Med. 2012;42:e3–28.

    Google Scholar 

  34. 34.

    Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc. 2000;32:963–75.

    CAS  Google Scholar 

  35. 35.

    Chen H, Cohen P, Chen S. How big is a big odds ratio? Interpreting the magnitudes of odds ratios in epidemiological studies. Commun Statistics Simul Comput. 2010;39:860–4.

    Google Scholar 

  36. 36.

    Cohen J. A power primer. Psychol Bull. 1992;112:155–9.

    CAS  Google Scholar 

  37. 37.

    Burnet D, Plaut A, Courtney R, Chin MH. A practical model for preventing type 2 diabetes in minority youth. Diab Educator. 2002;28:779–95.

    Google Scholar 

  38. 38.

    Schwarzer R. Modeling health behavior change: How to predict and modify the adoption and maintenance of health behaviors. Appl Psychol. 2008;57:1–29.

    Google Scholar 

  39. 39.

    Weinstein ND. The precaution adoption process. Health Psychol. 1988;7:355–86.

    CAS  Google Scholar 

  40. 40.

    Hall PA, Fong GT. Temporal self-regulation theory: A model for individual health behavior. Health Psychol Rev. 2007;1:6–52.

    Google Scholar 

  41. 41.

    Hagger M, Chatzisarantis NLD. An integrated behavior-change model for physical activity. Exerc Sport Sci Rev. 2014;42:62–9.

    Google Scholar 

  42. 42.

    Gӧhner W, Seelig H, Fuchs R. Intervention effects on cognitive antecedents of physical exercise: a 1-year follow-Up study. Applied Psychol Health Well-Being. 2009;1:233–56.

    Google Scholar 

  43. 43.

    Rhodes RE, de Bruijn GJ. What predicts intention-behavior discordance? A review of the action control framework. Exerc Sports Sci Rev. 2013;41(4):201–7.

    Google Scholar 

  44. 44.

    de Vries H, Mesters I, van de Steeg H, Honing C. The general public's information needs and perceptions regarding hereditary cancer: An application of the Integrated Change Model. Patient Educ Couns. 2005;56:154–65.

    Google Scholar 

  45. 45.

    Fisher JD, Fisher WA. Changing AIDS risk behavior. Psychol Bull. 1992;111:455–74.

    CAS  Google Scholar 

  46. 46.

    West R, Brown J. Theory of Addiction. 2nd ed. Oxford: Wiley-Blackwell; 2013.

    Google Scholar 

  47. 47.

    Baggozi RP. The poverty of economic explanations of consumption and an action theory alternative. Manag Decis Econ. 2000;21:95–109.

    Google Scholar 

  48. 48.

    Olander F, Thogersen J. Understanding of consumer behaviour as a prerequisite for environmental protection. J Consum Policy. 1995;18:345–85.

    Google Scholar 

  49. 49.

    Heckhausen H, Gollwitzer PM. Thought contents and cognitive functioning in motivational and volitional states of mind. Motiv Emot. 1987;11:101–20.

    Google Scholar 

  50. 50.

    Bagozzi RP. The self-regulation of attitudes, intentions, and behavior. Soc Psychol Q. 1992;55:178–204.

    Google Scholar 

  51. 51.

    Triandis HC. Interpersonal Behavior. Monterey: CABrooks/Cole; 1977.

    Google Scholar 

  52. 52.

    McAuley E, Courneya KS, Lettunich J. Effects of acute and long-term exercise on self-efficacy responses in sedentary, middle-aged males and females. The Gerontologist. 1991;31:534–42.

    CAS  Google Scholar 

  53. 53.

    West R. The multiple facets of cigarette addiction and what the mean for encouraging and helping smokers to stop. J Chronic Obsructive Pulm Dis. 2009;6:277–83.

    Google Scholar 

  54. 54.

    Valois P, Desharnai R, Godin G. A comparison of Fishbein and Ajzen and the Triandis attitudinal models for the prediction of exercise intention and behavior. J Behav Med. 1988;11:459–72.

    CAS  Google Scholar 

  55. 55.

    Blalock SJ, DeVellis RF, Giorgino KB, De Vellis BM, Gold DT, Dooley MA, et al. Osteoporosis prevention in premenopausal women: Using a stage model approach to examine the predictors of behavior. Health Psychol. 1996;15(2):84–93.

    CAS  Google Scholar 

  56. 56.

    Elliott JO, Seals BF, Jacobson MP. Use of the Precaution Adoption Process Model to examine predictors of osteoprotective behavior in epilepsy. Seizure. 2007;16:424–37.

    Google Scholar 

  57. 57.

    de Vries H, Kremers SPJ, Smeets T, Brug J, Eijmael K. The effectiveness of tailored feedback and action plans in an intervention addressing multiple behaviors. Am J Health Promot. 2008;22(6):417–25.

    Google Scholar 

  58. 58.

    Kelly S, Melnyk BM, Belyea M. Predicting physical activity and fruit and vegetable intake in adolescents: A test of information, motivation, behavioral skills model. Res Nurs Health. 2012;35:146–63.

    Google Scholar 

  59. 59.

    Osborn CY, Amico KR, Cruz N, O’Connell AA, Perez-Escamilla R, Kalichman SC, et al. A brief culturally tailored intervetion for Puerto Ricans with type 2 diabetes. Health Educ Behav. 2010;37:849–62.

    Google Scholar 

  60. 60.

    Gӧhner W, Schlatterer M, Seelig H, Frey I, Berg A, Fuchs R. Two-year follow-up of an interdisciplinary cognitive-behavioral intervention program for obese adults. J Psychol. 2012;146:371–91.

    Google Scholar 

  61. 61.

    Fuchs R, Seelig H, Gӧhner W, Burton NW, Brown WJ. Cognitive mediation of intervention effects on physical exercise: Causal models for the adoption and maintenance stage. Psychol Health. 2012;27:1480–99.

    Google Scholar 

  62. 62.

    Li K. Domain dimensionality and temporality of outcome expectancy for physical activity among middle-aged and older Chinese adults: A latent profile analysis. Psychol Sport Exerc. 2013;14:682–91.

    Google Scholar 

  63. 63.

    Barg CJ, Latimer AE, Pomery EA, Rivers SE, Rench TA, Prapavessis H, et al. Examining predictors of physical activity among inactive middle-aged women: An application of the health action process approach. Psychol Health. 2012;27:829–45.

    Google Scholar 

  64. 64.

    Berli C, Lorentini P, Radtke T, Hornung R, Scholz U. Predicting physical activity in adolescents: The role of compensatory health beliefs within the Health Action Process Approach. Psychol Health. 2014;29:458–74.

    Google Scholar 

  65. 65.

    Caudroit J, Stephan Y, Le Scanff C. Social cognitive determinants of physical activity among retired older individuals: An application of the health action process approach. Br J Health Psychol. 2011;16:404–17.

    Google Scholar 

  66. 66.

    Duan Y, Lippke S, Wagner P, Brehm W. Testing two stage assessments in a Chinese college student sample: Correspondences and discontinuity patterns across stages. Psychol Sport Exerc. 2011;12:306–13.

    Google Scholar 

  67. 67.

    Parschau L, Barz M, Richert J, Knoll N, Lippke S, Schwarzer R. Physical activity among adults with obesity: Testing the health action process approach. Rehabil Psychol. 2014;59:42–9.

    Google Scholar 

  68. 68.

    Renner B, Spivak Y, Kwon S, Schwarzer R. Does Age make a difference? predicting physical activity of south Koreans. Psychol Aging. 2007;22:482–93.

    Google Scholar 

  69. 69.

    Scholz U, Keller R, Perren S. Predicting behavioral intentions and physical exercise: A test of the health action process approach at the intrapersonal level. Health Psychol. 2009;28:702–8.

    Google Scholar 

  70. 70.

    Ziegelmann JP, Lippke S. Planning and strategy use in health behavior change: a life span view. Int J Behav Med. 2007;14:30–9.

    Google Scholar 

  71. 71.

    Chiu CY, Lynch RT, Berven NL. The health action process approach as a motivational model for physical activity self-management for people with multiple sclerosis: a path analysis. Rehabil Psychol. 2011;56:171–81.

    Google Scholar 

  72. 72.

    Kassavou A, Hamborg T, Turner A, French DP. Predicting maintenance of attendance of walking groups: Testing constructs from three leading maintenance theories. Health Psychol. 2014;33:752–6.

    Google Scholar 

  73. 73.

    Lippke S, Plotnikoff RC. Testing two principles of the health action process approach in individuals with type 2 diabetes. Health Psychol. 2014;33:77–84.

    Google Scholar 

  74. 74.

    Perrier MJ, Sweet SN, Strachan SM, Latimer-Cheung AE. I act, therefore I am: Athletic identity and the health action process approach predict sport participation among individuals with acquired physical disabilities. Psychol Sport Exerc. 2012;13:713–20.

    Google Scholar 

  75. 75.

    Ströbl V, Knisel W, Landgraf U, Faller H. A combined planning and telephone aftercare intervention for obese patients: Effects on physical activity and body weight after one year. J Rehabil Med. 2012;45:198–205.

    Google Scholar 

  76. 76.

    Lippke S, Schwarzer R, Ziegelmann JP, Scholz U, Schuz B. Testing stage-specific effects of a stage-matched intervention: A randomized controlled trial targeting physical exercise and its predictors. Health Educ Behav. 2010;37:533–46.

    Google Scholar 

  77. 77.

    Gaston A, Prapavessis H. Using a combined protection motivation theory and health action process approach intervention to promote exercise during pregnancy. J Behav Med. 2014;37:173–84.

    Google Scholar 

  78. 78.

    Sheeran P. Intention-behaviour relations: A conceptual and empirical review. In: Hewstone M, Stroebe W, editors. European Review of Social Psychology. Volume 12. Chichester, UK: John Wiley & Sons; 2002. p. 1–36.

    Google Scholar 

  79. 79.

    de Bruijn GJ. Exercise habit strength, planning and the theory of planned behaviour: An action control approach. Psychol Sport Exerc. 2011;12:106–14.

    Google Scholar 

  80. 80.

    de Bruijn GJ, de Groot R, van den Putte B, Rhodes RE. Conscientiousness, extroversion, and action control: comparing moderate and vigorous physical activity. J Sport Exerc Psychol. 2009;31:724–42.

    Google Scholar 

  81. 81.

    de Bruijn GJ, Verkooijen K, de Vries NK, van den Putte B. Antecedents of self identity and consequences for action control: An application of the theory of planned behaviour in the exercise domain. Psychol Sport Exerc. 2012;13:771–8.

    Google Scholar 

  82. 82.

    Fiala B, Rhodes RE. Understanding Exercise Action Control: Moderation of Intensity on Physical Activity Intentions and Behavior in a Canadian University Sample. Planned Behavior: Theory, Applications and Perspectives. Ch.5. Nova Publications. 2014;73.

  83. 83.

    Godin G, Shephard RJ, Colantonio A. The cognitive profile of those who intend to exercise but do not. Public Health Rep. 1986;101:521–6.

    CAS  Google Scholar 

  84. 84.

    Rhodes RE, Courneya KS, Jones LW. Translating exercise intentions into behavior: Personality and social cognitive correlates. J Health Psychol. 2003;8:447–58.

    Google Scholar 

  85. 85.

    Rhodes RE, Nasuti G, Fiala B. Action control of exercise behaviour: Evaluation of social cognition, cross-behavioural regulation and automaticity. Behav Med. 2012;38:121–8.

    Google Scholar 

  86. 86.

    Rhodes RE, Plotnikoff RC. Understanding action control: Predicting physical activity intention-behavior profiles across six months in a Canadian sample. Health Psychol. 2006;25:292–9.

    Google Scholar 

  87. 87.

    Rhodes RE, Plotnikoff RC, Courneya KS. Predicting the physical activity intention-behaviour profiles of adopters and maintainers using three social cognition models. Ann Behav Med. 2008;36:244–52.

    Google Scholar 

  88. 88.

    Rhodes RE, de Bruijn GJ, Matheson DH. Habit in the physical activity domain: Integration with intention temporal stability and action control. J Sport Exerc Psychol. 2010;32(1):84–98.

    Google Scholar 

  89. 89.

    Kaushal N, Rhodes RE. Exploring personality and physical environment as predictors of exercise action control. In: Psychology of Extraversion Perspectives in Psychology Research. New York, NY: Nova Science Publishers; 2014. p. 91–105.

    Google Scholar 

  90. 90.

    Rhodes RE, Naylor PJ, McKay HA. Pilot study of a family physical activity planning intervention among parents and their children. J Behav Med. 2010;33:91–100.

    Google Scholar 

  91. 91.

    Head KJ, Noar SM. Facilitating progress in health behaviour theory development and modification: the reasoned action approach as a case study. Health Psychol Rev. 2014;8:34–52.

    Google Scholar 

  92. 92.

    Schwarzer R. Life and death of health behaviour theories. Health Psychology Review 2013.

  93. 93.

    Rhodes RE. Improving translational research in building theory: a commentary on Head and Noar. Health Psychol Rev. 2014;8(1):57–60.

    Google Scholar 

  94. 94.

    Prestwich A, Sniehotta FF, Whittington C, Dombrowski SU, Rogers L, Michie S. Does theory influence the effectiveness of health behavior interventions? Meta-analysis. Health Psychol. 2014;33(5):465–74.

    Google Scholar 

  95. 95.

    Noar SM, Head KJ. Mind the gap: bringing our theories in line with the empirical data - a response to commentaries. Health Psychol Rev. 2013;8(1):65–9.

    Google Scholar 

  96. 96.

    Webb TL, Sheeran P. Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychol Bull. 2006;132:249–68.

    Google Scholar 

  97. 97.

    Carraro N, Gaudreau P. Spontaneous and experimentally induced action planning and coping planning for physical activity: A meta-analysis. Psychol Sport Exerc. 2013;14:228–48.

    Google Scholar 

  98. 98.

    Rhodes RE, Pfaeffli LA. Mediators of physical activity behaviour change among adult non-clinical populations: A review update. Int J Behav Nutr Phys Act. 2010;7:37.

    Google Scholar 

  99. 99.

    Kwasnicka D, Presseau J, White M, Sniehotta FF. Does planning how to cope with anticipated barriers facilitate health-related behaviour change? A systematic review. Health Psychology Review 2013, doi:10.1080/17437199.2013.766832.

  100. 100.

    Conn VS, Hafdahl AR, Mehr DR. Interventions to increase physical activity among healthy adults: Meta-analysis of outcomes. Am J Public Health. 2011;101:751–8.

    Google Scholar 

  101. 101.

    Bélanger-Gravel A, Godin G, Amireault S. A meta-analytic review of the effect of implementation intentions on physical activity. Health Psychology Review. 2013;7(1):1–32.

    Google Scholar 

  102. 102.

    Nigg CR, Geller KS, Motl RW, Horwath CC, Wertin KK, Dishman RK. A research agenda to examine the efficacy and relevance of theTranstheoretical Model for physical activity behavior. Psychol Sport Exerc. 2011;12:7–12.

    Google Scholar 

  103. 103.

    Sallis JF, Owen N. Ecological models. In: Glanz K, Lewis FM, Rimer BK, editors. Health Behavior and Health Education. San Francisco: Jossey-Bass; 1997. p. 403–24.

    Google Scholar 

  104. 104.

    Prins RG, van Empelen P, te Velde SJ, Timperio A, van Lenthe FJ, Tak NI, et al. Availability of sports facilities as moderator of the intention–sports participation relationship among adolescents. Health Educ Res. 2010;25:489–97.

    Google Scholar 

  105. 105.

    Rhodes RE, Brown SG, McIntyre CA. Integrating the perceived neighbourhood environment and the theory of planned behaviour when predicting walking in Canadian adult sample. Am J Health Promot. 2006;21:110–8.

    Google Scholar 

  106. 106.

    Rhodes RE, Courneya KS, Blanchard CM, Plotnikoff RC. Prediction of leisure-time walking: An integration of social cognitive, perceived environmental, and personality factors. Int J Behav Nutr Phys Act. 2007;4:51.

    Google Scholar 

  107. 107.

    Rhodes RE, Dickau L. Moderators of the intention-behavior relationship in physical activity: A systematic review. Br J Sports Med. 2013;47(4):215–25.

    Google Scholar 

  108. 108.

    Sheeran P, Gollwitzer PM, Bargh JA. Nonconscious processes and health. Health Psychol. 2013;32:460–73.

    Google Scholar 

  109. 109.

    Gardner B, de Bruijn GJ, Lally P. A systematic review and meta-analysis of applications of the Self-Report Habit Index to nutrition and physical activity behaviors. Ann Behav Med. 2011;42:174–87.

    Google Scholar 

  110. 110.

    Berry T, Strachan SM. Implicit and explicit exercise and sedentary identity. Res Q Exerc Sport. 2012;83:479–84.

    Google Scholar 

  111. 111.

    Oxford Dictionary. (2014). Oxford University Press.

  112. 112.

    Rhodes RE, Nasuti G. Trends and changes in research on the psychology of physical activity across 20 years: A quantitative analysis of 10 journals. Prev Med. 2011;53(1–2):17–23.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Ryan E Rhodes.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contribution

RER and CAY both carried out the literature search and participated in the coding and theming. RER wrote the main body of the paper while CAY wrote the Methods section and provided editorial feedback on the entire paper. Both authors read and approved the final manuscript.

Additional file

Additional file 1: Table S1.

Characteristics of physical activity studies that utilized models/frameworks/theories (n = 36).

Rights and permissions

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Rhodes, R.E., Yao, C.A. Models accounting for intention-behavior discordance in the physical activity domain: a user’s guide, content overview, and review of current evidence. Int J Behav Nutr Phys Act 12, 9 (2015).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Theory
  • Intention-behavior gap
  • Volition
  • Post-intention
  • Exercise
  • Action control