Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Oct 1.
Published in final edited form as: Pers Individ Dif. 2021 Jun 5;181:111000. doi: 10.1016/j.paid.2021.111000

You Are What You Repeatedly Do: Links Between Personality and Habit

Kiran McCloskey 1,1, Blair T Johnson 1
PMCID: PMC8562686  NIHMSID: NIHMS1709680  PMID: 34737484

Abstract

Several socio-cognitive theories of personality describe habit development as an integral process of personality development. Yet, no empirical research has rigorously examined linkages between personality traits and habits. In the present study, participants (N=459) reported perceived automaticity, intrinsic rewards, and extrinsic rewards for several of 25 total behaviors; they also self-reported two important traits: conscientiousness and neuroticism. Conditional multilevel mediation analyses were used to assess the effect of each conscientiousness and neuroticism on automaticity through intrinsic and extrinsic rewards for each behavior. Across behaviors, conscientiousness was negatively associated with behavioral automaticity, and neuroticism positively predicted it. Specifically, conscientiousness appeared to protect against automaticity for health risk behaviors but did not promote automaticity for behaviors performed frequently by those high in conscientiousness; conversely, neuroticism positively predicted automaticity even for behaviors not performed more frequently by those high on the trait. Perceived intrinsic and extrinsic rewards mediated the link between traits and automaticity for some behaviors (e.g., sugary drink consumption), but these effects were not consistent across all behaviors. These findings offer some of the first empirical insights into the links between personality and habits.

Keywords: personality, habit, conscientiousness, neuroticism, behavior

1. Introduction

Personality has been frequently described as “relatively enduring patterns of thinking, feeling, and behavior” (Roberts & Mroczek, 2008, p. 31), whereas habits are repeated behaviors executed automatically in response to particular environmental cues (Lally & Gardner, 2013). Thus, ‘personality’ and ‘habit’ each refer to patterns of behavior, but personality reflects more global, general patterns, and habit describes specific cue-behavior associations. Despite this distinction, theory has suggested the two constructs and their processes may be interrelated (e.g. Wrzus & Roberts, 2017). Yet, the associations between habits and personality have not been rigorously empirically tested; the present study addresses this gap.

1.1. Habit and Personality: Theoretical Parallels

Habits are broad and pervasive; it has been suggested that up to 45% of our daily behavior can be classified as habitual (Wood, Quinn, & Kashy, 2002). Specifically, habits are learned cue-behavior associations, in which repeated execution of a behavior in the presence of an environmental cue comes to automatically yield an “impulse” toward that behavior when that cue is encountered again in the future (Lally & Gardner, 2013), especially so if that behavior is rewarded (Gardner & Lally, 2018).

Personality literature often runs parallel to concepts in habits research. In line with the reinforcement learning and automaticity themes of habits literature, social-cognitive approaches to personality have posited that personality develops as a result of individuals’ learned behavior-outcome experiences, as well as individuals’ values of such outcomes. Two such frameworks of personality development, the TESSERA framework (Wrzus & Roberts, 2017), and the self-regulated framework of personality change (Hennecke, Bleidorn, Denissen, & Wood, 2014) posit that personality shifts as new behaviors are rewarded and become automatic, and even describe habits by name as an integral process by which personality development processes. Specifically, short-term states are behavioral, cognitive, or affective conditions that individuals experience as a result of a triggering situation (Hennecke et al., 2014; Wrzus & Roberts, 2017); when new states become habitual or automatic, personality shifts.

Despite the plausibility of this connection, to date, the potential associations between personality and habits have not been examined in-depth. Personality-behavior research has primarily studied associations between traits and behavioral frequency only (e.g., Bogg & Roberts, 2004). The TESSERA and self-regulated frameworks for personality change describe automatic habits only in passing and without detailed reference to habits literature. Similarly, when the habits literature has considered personality, it has largely only done so by assessing if a particular trait is correlated with habits or automaticity for a specific behavior. This line of research has yielded mixed results (Judah, 2015; Vishwanath, 2015; Wood et al., 2002), which is unsurprising; according to the Principle of Compatibility, constructs best predict behaviors when the predictor and the behavior match in terms of specificity (Ajzen & Fishbein, 2005). As personality is a general construct, it can thus be predicted that personality will best predict behavior when behavior is assessed generally as well.

1.2. Rewards in Habit and Personality Processes

If there is an association between personality and behavioral automaticity across behaviors, one mechanism through which personality may influence habits is through perceived rewards. The corresponsive principle of personality stability posits that life experience will reinforce the traits that caused a person to select their circumstances in the first place (Caspi, Roberts, & Shiner, 2005). For instance, a conscientious individual may choose to attend graduate school, where the academic environment further rewards detail-oriented and organized behaviors. Personality may also influence perceptions of rewards; previous theory has posited that completing an intended behavior is inherently rewarding (Lally & Gardner, 2013), and conscientious individuals may feel particularly rewarded by executing an intended action. Conversely, neurotic individuals may be more likely to develop tension reduction expectancies, which may enhance their perceptions of rewards for negatively-reinforcing behaviors (Wiers, 2008).

Behaviors may be rewarding if they yield an outcome an individual evaluates as favorable. These rewards may be extrinsic, helping the person achieve a desired goal, or these rewards may be intrinsic if that behavior yields pleasure or positive feelings regardless of other outcomes. Although both extrinsic and intrinsic rewards have the theoretical potential to yield habit development, habit literature has argued that intrinsic rewards have a stronger impact on habits than extrinsic rewards (Gardner & Lally, 2018); thus, if personality influences automaticity through rewards, this pathway is more likely to occur through intrinsic rewards than extrinsic rewards.

1.3. The Present Study

Two traits, conscientiousness and neuroticism, are particularly relevant to understanding habits. On the one hand, conscientious individuals tend to be more self-controlled and orderly, and such tendencies may provide an advantage during early habit formation processes which require behavior initiation and repetition (Judah, 2015). On the other hand, neuroticism is also of particular interest, as this trait has been associated with several behaviors that might be considered habitual, such as alcohol consumption (Malouff, Thorsteinsson, Rooke, & Schutte, 2007) and dependence on e-cigarettes (Zvolensky, Shepherd, Garey, Case, & Gallagher, 2020). Such behaviors may partially account for the associations between neuroticism and physical and health (Lahey, 2009).

The present research thus explores the potential associations between conscientiousness, neuroticism, and habits, and examines trends across a broad spectrum of 25 behaviors. We predict that behaviors frequently performed by people high on conscientiousness or neuroticism will also be more automatic for individuals high on those traits, respectively. The association between traits and automaticity will also be greatest for behaviors that are viewed as rewarding for individuals high on the same traits, and intrinsic rewards, rather than extrinsic rewards, will drive this pathway.

2. Method

The present study relies on the data McCloskey and Johnson (2019) collected, which is available for secondary use. Although data collection procedures and preliminary analyses were described in-depth in the previous publication, the most important details for the current study are reiterated briefly here.

2.1. Participants and Procedures

In total, 459 adults residing within the United States were recruited via MTurk to fill out a survey on daily behavior. Each participant was randomized into one of three groups to reduce participant demand; in each group, participants rated 11 behaviors, of which seven behaviors were exclusive to that group, and four behaviors were rated by all participants in all groups: exercise, smoking, handwashing, and medication adherence. In total, this study collected ratings on 25 different behaviors: the other 21 were alcohol consumption, unhealthy snacking, fruit and vegetable consumption, active commuting, making savings deposits, playing music, texting and driving, checking one’s phone, food safety practices, playing videogames, sugary drink consumption, recycling, flossing, internet use, condom use, IT use, car use, seafood consumption, sunscreen use, sitting, and negative self- thoughts. These 25 behaviors represent behaviors that have been previously assessed in the habits literature, as identified by a previous meta-analytic investigation (Low, 2016). The protocol was approved by the University of Connecticut Institutional Review Board (IRB) on August 9, 2018 (Protocol #X18–095).

2.3. Measures

2.3.1. Behavior Level (Level-1) Measures

2.3.1.1. Behavior Engagement.

Participants indicated the degree to which they performed each behavior on a 7-point Likert scale as a qualifier question (“To what extent do you [do behavior]?”). If participants did “not at all” engage in a specific behavior, they were not directed to provide ratings on any further measures used in this analysis.

2.3.1.2. Intrinsic Rewards.

Intrinsic rewards were assessed using a single item: “When I [do behavior] it is pleasurable”, rated on a 10-point scale.

2.3.1.2. Extrinsic Rewards.

Extrinsic rewards were assessed as perceived behavioral outcomes using two items: “To do [behavior] is good for me” and “I do [behavior] in order to achieve a goal”, rated on 7-point Likert scales. This measure showed good reliability in the present sample (α=.76).

2.3.1.2. Automaticity.

Automaticity was assessed using the four-items (e.g., “[Behavior] is something that I do without thinking”) from the Self-Reported Behavioral Automaticity Index (SRBAI: Gardner, Abraham, Lally, & de Brujin, 2012), each measured on a 7-point Likert scale.. Present reliability for this measure was high (α=.96).

2.3.2. Participant level (Level-2) variables

2.3.2.1. Personality.

Conscientiousness and neuroticism were assessed with four items each from the Mini-IPIP (Donnellan, Oswald, Baird, & Lucas, 2006), which is a popular, validated short assessment of the Big Five personality traits. Each item was rated on a 5-point scale. Reliability in the present sample was acceptable: (for conscientiousness: α=0.72; for neuroticism: α=0.76).

2.4. Analyses

All variables were transformed to percent of maximum possible (POMP) scores; POMP scoring eases comparison across varying scales with differing, often arbitrary units (Cohen, Cohen, Aiken, & West, 1999). Conditional mediation analyses assessed the influence of traits on automaticity through each extrinsic and intrinsic rewards for each behavior (Figure 1). By including behavior as a dummy-coded covariate in a single model and estimating effects at each level of the covariate simultaneously, the issues involved with multiple comparisons were mitigated. In order to account for repeated measurements within participants, multilevel modelling was utilized in which participant was included as a random effect on intercepts. Random slopes were not included because the data did not include multiple observations for all participants, and thus the random slopes model failed to converge.

Figure 1.

Figure 1.

Open in a new tab

Conceptual total mediation model across behaviors.

Note. The circle around the dependent variable represents random intercepts between participants.

To describe trends across behaviors, total and direct effects for all behaviors were explored graphically using ggplot2 (Wickham, 2016). Models were run separately for each trait to independently test the effects of conscientiousness and neuroticism. Multilevel models were run using the lme4 package in R (Bates, Maechler, Bolker, & Walker, 2015), and mediation analyses were conducted using the mediation package (Tingley, Yamamoto, Hirose, Keele, & Imai, 2014). All analyses were conducted using version 3.5.1 of R.

3. Results

On average, each participant provided ratings for 8 behaviors, and each behavior was rated, on average, by 152 participants, making for a total of 3,950 total behavior observations. Recruited participants tended to be relatively conscientious (M=73.89, SD=17.87) and scored in the midranges of neuroticism (M=50.11, SD=20.08). The recruited sample showed similar demographic breakdowns to the typical MTurk sample (Huff & Tingley, 2015). Full demographic statistics appear in the supplemental materials (Table S1). The intraclass correlation score for automaticity was ICC=.21.

3.1. Trait Associations

Preliminary analyses showed that conscientiousness was significantly associated with higher reported behavioral engagement for fruit and vegetable consumption (β=.176, p=.03, 95% CI[.16, .19]), food safety practices (β=.385, p<.001, 95% CI[.374, .395]), and handwashing (β=.169, p<.001, 95% CI[.16, .18.]); it was associated with lower rates of smoking (β=−.149, p=.04, 95% CI[−.16, −.14]), sitting (β=−.259, p=.001, 95% CI[−.27, −.25]), taking medication (β=−.221, p<.001, 95% CI [−.23, −.21]), sugary drink consumption (β=−.222, p=.02, 95% CI[−.24, −.21]), unhealthy snacking (β=−.298, p<.001, 95% CI[−.31, −.28]), playing music (β=−.220, p=.03, 95% CI[−.24, −.20], texting and driving (β=−.478, p<.001, 95% CI[−.50, −.45] and negative self-thoughts (β=−.494, p<.001, 95% CI[−0.51, −0.48]). Neuroticism was significantly associated with higher rates of taking medication (β=.306, p<.001, 95% CI[.30, .32]), sitting (β=.254, p=.002, 95% CI[.25, .26], and negative self-thoughts (β=.632, p<.001, 95% CI[.62, .64]), and significantly associated with lower rates of exercise (β=−.185, p<.001, 95% CI [−.19, −.18]). Figure 2 shows the associations between each behavior and conscientiousness and neuroticism; exact scores are available in the supplemental materials (Table S2). Neuroticism and conscientiousness were themselves inversely correlated r=−.41, p<.001 (Table 2), and Figure 2 suggests that the associations between each behavior with each conscientiousness and neuroticism are also often negatively correlated.

Figure 2.

Figure 2.

Open in a new tab

Correlations between behavioral engagement and conscientiousness (x) and neuroticism (y).

Note. All labeled behaviors show significant associations between behavioral engagement and at least one trait. The bold black line shows the best-fitting linear line of the two effects.

Table 2.

Sample-size weighted between-participant correlations.

1. 2. 3. 4. 5.
1. Conscientiousness —
2. Neuroticism −0.417*** —
3. Automaticity −0.133** 0.123** —
4. Frequency −0.134** 0.124** 0.401*** —
5. Extrinsic rewards 0.105* −0.085 0.354*** 0.349*** —
6. Intrinsic rewards −0.023 −0.002 0.402*** 0.210*** 0.349***
Open in a new tab

Note.

*

p<.05

**

p<.01

***

p<.001.

3.2. Main Analyses

3.2.1. Conscientiousness

Model fit showed that the mixed-effects model significantly outperformed a fixed-effects model, χ2(1, N=459)=208.33, p<.001 Results from the mixed-effects model showed that across behavior categories, conscientiousness was negatively associated with automaticity (β=−0.06, p=.007, 95% CI[−.18, −.01]). Conversely, both intrinsic (β=0.13, p<.001, 95% CI[.10, .17]: Figure 1, path b1) and extrinsic rewards (β=0.08, p<.001, 95% CI[.05, .11]: Figure 1, path b2) were positively associated with automaticity across behaviors. The total effect of conscientiousness on automaticity (Figure 1, path c) was significant and positive for food safety practices (β=.20, p=.04, 95% CI[.002, .45]) but significantly negative for exercise (β=−.16, p<.001, 95% CI[−.30, −.07]), internet use (β=−.23, p=.04, 95% CI[−.47, −.01]), alcohol consumption (β=−.35, p<.001, 95% CI[−.62, −.14]), seafood consumption (β=−.31, p<.001, 95% CI[−.50, −.09]), playing video games (β=−.29, p<.001, 95% CI[−.53, −.07]), playing music (β=−.49, p<.001, 95% CI[−.83, −.25]), texting and driving (β=−.60, p<.001, 95% CI[−.98, −.22]), sugary drink consumption (β=−.26, p<.001, 95% CI[−.47, −.03]), unhealthy snacking (β=−.45, p<.001, 95% CI[−.67, −.27]), and negative self-thoughts (β=−.24; p=.04, 95% CI[−.48, −.02]). As Figure 3 demonstrates, when conscientiousness was negatively associated with behavioral engagement, conscientiousness was also often negatively associated with behavioral automaticity (Figure 1, path c). Yet, only one behavior with a positive association between behavior engagement and conscientiousness – food safety – also showed a significant association between automaticity and conscientiousness.

Figure 3.

Figure 3.

Open in a new tab

Total effect of conscientiousness on automaticity (path c) by behavior (y) as a function of the degree of association between conscientiousness and behavioral engagement (x).

Note. All labeled behaviors show significant total effects of conscientiousness on automaticity. The bold black line shows the best-fitting linear line of the two effects.

Of the 11 behaviors that showed significant total effects of conscientiousness on automaticity, six also showed significant indirect effects by which rewards mediated the link between trait and automaticity. Extrinsic rewards partially mediated the associations between conscientiousness and automaticity for sugary drink consumption (β=−.04, p<.001, 95% CI[−.07, −.02]) and unhealthy snacking (β=−.04, p<.001, 95% CI[−.07, −.02]). Intrinsic (β=−.19, p<.001, 95% CI[−.27, −.10]) and extrinsic rewards (β=−.06, p<.001, 95% CI[−.12, −.02]) together both partially mediated the link between conscientiousness and automaticity of texting and driving. After accounting for the indirect effect of intrinsic (β=−.08, p<.001, 95% CI[−.14, −.03]) and extrinsic rewards (β=−.04, p<.001, 95% CI[−.07, −.02]), the effect of conscientiousness on automaticity of negative self-thoughts was no longer significant. Interestingly, when the indirect effects of extrinsic (β=−.02, p<.001, 95% CI[.01, .04]) and intrinsic rewards (β=.05, p<.001, 95% CI[.01, .08]) were included in the mediation model for exercise, the direct effect of conscientiousness on automaticity for exercise was stronger than the total effect of conscientiousness on exercise automaticity, indicating a suppression effect of rewards on exercise (Tzelgov & Henik, 1991). Supplemental Table S3 presents the results of the mediation model for each of the 25 behaviors, ordered by decreasing behavioral association with conscientiousness.

3.2.2. Neuroticism

Model fit showed that the mixed-effects model significantly outperformed a fixed-effects model, χ2(1, N=459)=207.81, p<.001. Results from the mixed-effects model showed that across behaviors, neuroticism was positively associated with automaticity (β=0.08, p=.007, 95% CI[0.01, 0.16]); the effects of intrinsic (Figure 1, path b1) and extrinsic rewards (Figure 1, path b2) on automaticity were equivalent to those estimated from the models focused on conscientiousness. The total effect of neuroticism on automaticity (Figure 1, path c) was significant and positive for negative self-thoughts (β=.40, p<.001, 95% CI[.17, .62]), sugary drink consumption (β=.27, p=.02, 95% CI[.02, .45]), unhealthy snacking (β=.22, p<.001, 95% CI[.02, .43]), IT use (β=.21, p=.02, 95% CI[.03, .32]), internet use (β=.35, p<.001, 95% CI[.19, .55]), fruit and vegetable consumption (β=.20, p=.02, 95% CI[.06, .45]), playing video games (β=.20, p=.02, 95% CI[.02, .44]), and seafood consumption (β=.21, p=.02, 95% CI[.05, .48]); of these, only negative self-thoughts also showed a significant association between neuroticism and behavioral engagement (Figure 4). Only one behavior, sugary drink consumption, showed a significant total effect of neuroticism on automaticity that was mediated by either reward variable: greater perceived intrinsic rewards mediated the link between neuroticism and automaticity (β=.03, p<.001, 95% CI[.01, .06]) for this behavior. Table S4 presents the full mediation effects of intrinsic and extrinsic rewards on the link between neuroticism and automaticity, for each of the 25 measured behaviors.

Figure 4.

Figure 4.

Open in a new tab

Total effect of neuroticism on automaticity (path c) by behavior (y) as a function of the degree of association between neuroticism and behavioral engagement (x).

Note. All labeled behaviors show significant total effects of neuroticism on automaticity. The bold black line shows the best-fitting linear line of the two effects.

4. Discussion

The present study explored the association between traits and habits by examining the link between personality and habit on aggregate across multiple behaviors. The first hypothesis predicted that the behaviors with an association between traits and behavioral engagement would also show an association between the same trait and automaticity. In partial but weak support of this hypothesis, several risk behaviors (e.g., sugary drink consumption, unhealthy snacking) that were performed less frequently by individuals high on conscientiousness were also rated as less automatic by conscientious individuals; neurotic individuals also engaged in negative self-thoughts more frequently and rated this behavior as more automatic. Yet, these patterns were not fully consistent across behaviors. The second hypothesis predicted that intrinsic and extrinsic rewards would mediate the associations between traits and automaticity, and this effect did emerge for some behaviors (e.g., texting and driving, sugary drink consumption), but not all (e.g., food safety practices, alcohol consumption). The third hypothesis stated that intrinsic rewards, rather than extrinsic rewards, would account for the link between traits and automaticity. This hypothesis was not supported; more behaviors showed links between traits and automaticity mediated by extrinsic rewards than by intrinsic rewards.

By assessing patterns across 25 behaviors, the present study offers insights that assessing a smaller sample of behaviors would not. For instance, previous studies have assessed the association between personality traits and specific behaviors and have concluded that personality does not predict habit (Judah, 2015; Wood et al., 2002). Yet, these studies have not focused on health risk behaviors. In the present study, although the automaticity of health promotion behaviors largely was not predicted by either trait, health risk behaviors such as sugary drink consumption, texting and driving, and unhealthy snacking showed significant associations between automaticity and traits. The present findings suggest that conscientiousness may provide a protective health effect by preventing against the development of automaticity for these risky behaviors, whereas neuroticism may foster automaticity for health risk behaviors. This pattern aligns with previous meta-analytic evidence that conscientiousness more strongly predicts abstaining from health risk behaviors than it does predict engaging in health-promoting behaviors (Bogg & Roberts, 2004).

A number of surprising results deserve further consideration. First, the finding that extrinsic rewards – rather than intrinsic rewards – were more frequently a significant mediator of the associations between traits and automaticity suggests that extrinsic rewards may be more important in habit processes than previously thought (e.g., Gardner & Lally, 2018). The present findings also do not disprove the importance of intrinsic rewards; this study suggests that future research should consider both intrinsic and extrinsic rewards in order to examine the associations between traits and habits. Nevertheless, future research might benefit from assessing more specific rewards. Negative reinforcement might be considered a form of intrinsic rewards, but in the present study, intrinsic rewards were only assessed as the extent to which a behavior was perceived as ‘pleasurable’, not the extent to which a behavior relieved a negative experience. As previous literature has suggested neurotic individuals may be more sensitive to negative reinforcement (Wiers, 2008), future literature may find stronger effects of intrinsic rewards by considering negative reinforcement as well. Additionally, as extrinsic rewards were measured as the extent to which behaviors were perceived as “good for” the participant, extrinsic rewards in this study inherently conferred a longer-term benefit compared to the immediate intrinsic reward of pleasure. Thus, the links between traits and behavioral automaticity in this study may often be mediated by lasting, long-term rewards relative to immediate rewards, and future research may be needed to disentangle the immediacy of rewards from their locus.

Further, conscientiousness and neuroticism showed different patterns of automaticity, by which conscientiousness tended to be associated with lower automaticity, whereas neuroticism tended to be associated with higher automaticity. Previous personality development theory that has described habit development as an important aspect of personality development processes (e.g., Hennecke et al., 2014; Wruz & Roberts, 2017) generalizes personality development on aggregate, and has rarely differentiated personality processes by trait. Yet, the present results suggest that associative processes such as habit development may play a greater role for some traits (e.g., neuroticism) than others (e.g., conscientiousness). Conscientiousness may be a trait defined by a lack of automaticity, which is unsurprising given that deliberation has been described as a facet of conscientiousness (Costa & McCrae, 1992). Additionally, only one mediation effect through rewards emerged for neuroticism, suggesting that neuroticism might influence automaticity through alternate pathways; one potential pathway may be through greater experiences of stress, which shift goal-directed action to habitual control (Vogel et al., 2016)

Although it is not feasible in the current article to discuss all behaviors individually, the suppression effect of rewards on the association between conscientiousness and exercise automaticity merits further attention. In this study, the total effect of conscientiousness on exercise automaticity was significant and negative; after accounting for reward variables, the direct effect was stronger. As conscientiousness predicted higher perceptions of intrinsic and extrinsic rewards for exercise, and that intrinsic and extrinsic rewards were positively associated with automaticity, this finding suggests that conscientiousness itself is associated with weaker exercise automaticity, but that this effect is reduced when conscientious individuals perceive exercise to be more rewarding. This finding further supports the conclusion that conscientious individuals engage in behaviors – either healthful or risky – with more deliberation, but that perceptions of rewards nevertheless promote automaticity.

4.1. Limitations and Future Directions

Although the present study assessed many more behaviors than typically assessed in either habit or personality research, this study is limited in that patterns across behaviors were not examined statistically; trends were only summarized after visual examination. This method was adopted to avoid issues with statistical analysis and interpretation. As the associations between behavioral engagement and personality traits were estimates and many were not significantly different from zero, using these associations as true scores for statistical analysis would be inappropriate. As a result, the conclusions that can be drawn from the general trends are only tentative.

Additionally, as the present study focuses on a correlational, cross-sectional sample, directional effects cannot be identified; theory has both considered personality as a precursor of habit development and habit as a precursor of personality development, and longitudinal replication is required to identify the directionality of potential trait-habit associations, or if such associations are bidirectional. Because this study also relies on self-report measures of personality, behavior, and habit, participant self-perception biases and ability to report accurately on their experiences may also limit the findings. For instance, as participants self-reported automaticity, conscientious individuals may be inclined to report engaging in behaviors with careful, deliberate thought; this may create an illusion of lower automaticity for conscientious individuals regardless of actual habit strength.

Measures used in this study were also relatively short (four items or less). While these measures were adopted to reduce participant demand when assessing so many behaviors, this may introduce issues of measurement validity. Additionally, the short form scales of personality have may underestimate the role of personality traits in influencing behavior. In the present study, items measuring conscientiousness tended to be focused on the tendency to be organized and orderly; effects of conscientiousness on automaticity may be even stronger if measures focused on the facets of self-control or deliberativeness (Costa & McCrae, 1992).

It should also be noted that in Ryan and Deci’s (2000) conceptualization of intrinsic and extrinsic motivation, extrinsic motivation is divided into four types: external, introjected, identified, and integrated. In the present study, our measure of extrinsic rewards maps most closely onto identified or integrated motivation, which may not be as distinct from intrinsic motivation as external or introjected forms of extrinsic motivation. Thus, rewards in the present study may not be sufficiently disparate to identify separate pathways for intrinsic and extrinsic rewards; this is further evidenced by the relatively high correlation between the two reward measures (r=.35). Future analyses should extend consideration of extrinsic rewards to introjected and external rewards in order to fully map the differences between intrinsic and extrinsic rewards in habit and personality processes.

4.2. Conclusions

In the current study, personality does have some links to automaticity across behaviors, but these associations are dependent on both the trait and the behaviors being assessed. Conscientiousness tends to protect against automaticity for health risk behaviors, whereas neuroticism predicts greater automaticity for health risk behaviors as well as other behaviors. Conscientiousness, more so than neuroticism, predicts both extrinsic and intrinsic rewards for some behaviors, but these rewards do not fully mediate the associations between either trait and automaticity. These findings provide direction for future research, both in the personality and the habits domain, by emphasizing the differential mechanisms underlying the processes of different traits, bolstering the importance of extrinsic as well as intrinsic attitudes, and lending support for the consideration of habit prevention for risk behaviors. The present study thus bridges the gap between two literatures and lays the groundwork for further collaboration between two disparate, but related, lines of research.

Supplementary Material

1

Highlights.

  • Conscientiousness tends to predict lower automaticity for health risk behaviors

  • Neuroticism tends to predict higher automaticity for several behaviors

  • Perceived intrinsic and extrinsic rewards did not consistently mediate the links between traits and automaticity.

5. Acknowledgements

We thank JooChul Lee and Dr. Timothy E. Moore for their advice on statistical analyses and interpretations thereof; we also thank two anonymous reviewers for their input.

This study was supported, in part, by the U.S. National Institutes of Health (NIH) Science of Behavior Change Common Fund Program through an award administered by the National Institute on Aging (5U24AG052175). Kiran McCloskey was also supported by the Jorgensen Fellowship at the University of Connecticut during the development of this study.

Footnotes

CRediT Statement

Kiran McCloskey: Conceptualization, Methodology, Formal Analysis, Writing - Original Draft, Visualization, Writing - Reviewing and Editing. Blair T. Johnson - Conceptualization, Methodology, Visualization, Writing - Reviewing and Editing, Supervision.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Ajzen I, & Fishbein M. (2005). The influence of attitudes on behavior. In Albarracín D, Johnson BT, and Zanna MP (Ed.), The handbook of atttitudes (pp. 173–221). New York: Psychology Press. [Google Scholar]
  2. Bates D, Maechler M, Bolker B. & Walker S. (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1–48. doi: 10.18637/jss.v067.i01. [DOI] [Google Scholar]
  3. Bogg T, & Roberts BW (2004). Conscientiousness and health-related behaviors: A meta-analysis of the leading behavioral contributors to mortality. Psychological Bulletin, 130(6), 887–919. [DOI] [PubMed] [Google Scholar]
  4. Caspi A, Roberts BW, & Shiner RL (2005). Personality development: Stability and change. Annual Review of Psychology, 56, 453–484. doi: 10.1146/annurev.psych.55.090902.141913 [DOI] [PubMed] [Google Scholar]
  5. Cohen P, Cohen J, Aiken LS, & West SG (1999). The problem of units and the circumstance for POMP. Multivariate Behavioral Research, 34(3), 315–346. [Google Scholar]
  6. Costa PT, & McCrae RR (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five Factor Inventory (NEO-FFI): Professional manual. Psychological Assessment Resources, Inc. [Google Scholar]
  7. Donnellan MB, Oswald FL, Baird BM, & Lucas RE (2006). The mini-IPIP scales: Tiny-yet-effective measures of the big five factors of personality. Psychological Assessment, 18(2), 192–203. doi: 10.1037/1040-3590.18.2.192 [DOI] [PubMed] [Google Scholar]
  8. Gardner B, Abraham C, Lally P, & de Bruijn GJ (2012). Towards parsimony in habit measurement: Testing the convergent and predictive validity of an automaticity subscale of the Self-Report Habit Index. International Journal of Behavioral Nutrition and Physical Activity, 9(1), 102. doi: 10.1186/1479-5868-9-102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Gardner B. & Lally P. (2018) Modeling habit formation and its determinants. In Verplanken B. (Ed.), The psychology of habit (pp. 207–229). London, UK: Springer. [Google Scholar]
  10. Hennecke M, Bleidorn W, Denissen JJA, & Wood D. (2014). A three-part framework for self-regulated personality development across adulthood. European Journal of Personality, 28(3), 289–299. doi: 10.1002/per.1945 [DOI] [Google Scholar]
  11. Huff C, & Tingley D. (2015). “Who are these people?” Evaluating the demographic characteristics and political preferences of MTurk survey respondents. Research & Politics, 2(3), 1–12. doi: 10.1177/2053168015604648 [DOI] [Google Scholar]
  12. Judah GD (2015). An investigation into the psychological determinants of health habit formation. (Unpublished Ph.D.). University of London, London. 10.17037/PUBS.02121556 [DOI] [Google Scholar]
  13. Lahey BB (2009). Public health significance of neuroticism. American Psychologist, 64(4), 241–256. doi: 10.1037/a0015309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lally P, & Gardner B. (2013). Promoting habit formation. Health Psychology Review, 7, S137–S158. doi: 10.1080/17437199.2011.603640 [DOI] [Google Scholar]
  15. Low R. (2016). Place in habits and habits in place (Doctoral dissertation, University of Connecticut). Retrieved from https://opencommons.uconn.edu/dissertations/1103/. [Google Scholar]
  16. Malouff JM, Thorsteinsson EB, Rooke SE, & Schutte NS (2007). Alcohol involvement and the five-factor model of personality: A meta-analysis. Journal of Drug Education, 37(3), 277–294. [DOI] [PubMed] [Google Scholar]
  17. McCloskey K, & Johnson BT (2019). Habits, quick and easy: perceived complexity moderates the associations of contextual stability and rewards with behavioral automaticity. Frontiers in Psychology, 10, 1556. doi: 10.3389/fpsyg.2019.01556 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Roberts BW, & Mroczek D. (2008). Personality trait change in adulthood. Current Directions in Psychological Science, 17(1), 31–35. doi: 10.1111/j.1467-8721.2008.00543.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ryan RM, & Deci EL (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25(1), 54–67. [DOI] [PubMed] [Google Scholar]
  20. Tingley D, Yamamoto T, Hirose K, Keele L, & Imai K. (2014). mediation: R Package for Causal Mediation Analysis. Journal of Statistical Software, 59(5), 1–38. doi: 10.18637/jss.v059.i0526917999 [DOI] [Google Scholar]
  21. Tzelgov J, & Henik A. (1991). Suppression situations in psychological research: Definitions, implications, and applications. Psychological Bulletin, 109(3), 524–536. [Google Scholar]
  22. Vishwanath A. (2015). Examining the distinct antecedents of e-mail habits and its influence on the outcomes of a phishing attack. Journal of Computer-Mediated Communication, 20(5), 570–584. doi: 10.1111/jcc4.12126 [DOI] [Google Scholar]
  23. Vogel S, Fernández G, Joëls M, & Schwabe L. (2016). Cognitive adaptation under stress: a case for the mineralocorticoid receptor. Trends in Cognitive Sciences, 20(3), 192–203. doi: 10.1016/j.tics.2015.12.003 [DOI] [PubMed] [Google Scholar]
  24. Wickham H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag. [Google Scholar]
  25. Wiers RW (2008). Alcohol and drug expectancies as anticipated changes in affect: negative reinforcement is not sedation. Substance Use & Misuse, 43(3–4), 429–444. doi: 10.1080/10826080701203021 [DOI] [PubMed] [Google Scholar]
  26. Wood W, Quinn JM, & Kashy DA (2002). Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 83(6), 1281–1297. doi: 10.1037/0022-3514.83.6.1281 [DOI] [PubMed] [Google Scholar]
  27. Wrzus C, & Roberts BW (2017). Processes of personality development in adulthood: The TESSERA framework. Personality and Social Psychology Review, 21(3), 253–277. [DOI] [PubMed] [Google Scholar]
  28. Zvolensky MJ, Shepherd JM, Garey L, Case K, & Gallagher MW (2020). The influence of neuroticism in terms of E-cigarette dependence and beliefs about use and quitting among dual users of combustible and electronic cigarettes. Addictive behaviors, 107, 106396. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1709680-supplement-1.docx (31.5KB, docx)

RESOURCES