Correlations were specified between residuals that share common variance because they were created by the same first-order effect indicators. Advantages of this approach are stable model estimates, the availability of fit measures, and the robustness in terms of statistical assumptions for more details, see [ 61 ]. To report and evaluate all estimated models, empirically based recommendations were considered [ 62 , 63 ].
Therefore, multiple indices were used including both incremental fit indices that compare research models with a baseline model assuming independence of all variables and residual-based indices that evaluate the amount of error of model estimation [ 59 , 63 ]. All models were estimated based on the covariance matrix; easier interpretable correlations between all indicator variables are displayed in Table A in S1 Appendix. This study was carried out in accordance with the Declaration of Helsinki.
All data were collected during a study on the interrelation of traits related to NFC and Self-Control that was originally planned as preliminary study. That study was initially approved by the local data security administrator. All coefficients indicated comparably good to high reliability.
Therefore, models were estimated using maximum likelihood estimation with robust standard errors.
Cognition and action
For analyzing latent interactions, mean- and variance-adjusted maximum likelihood MLMV was used. To verify the quality of the parceling method, average factor loadings of the items on the respective factor were calculated per parcel. The average loadings were at least. Then, we determined the measurement model of Self-Control.
In this, we tested two models. Following our research interest to examine general Self-Control, the first parsimonious model assumed that all indicators of both instruments to assess Self-Control loaded on one common factor. In comparison, a second-order model took into account that while both instruments focus on different behavioral aspects of Self-Control [ 56 , 58 ], they nevertheless assess a common core construct. It assumed two first-order factors corresponding to both instruments and one second-order factor reflecting the shared variance due to a common Self-Control factor see Fig 2.
This model was more complex, but theoretically favored because it postulates a general Self-Control factor while also representing a substructure that refers to different research traditions and corresponding differences in the specific Self-Control behavior assessed. The hierarchical factor model was chosen instead of a correlated factor model because we were interested in relations of NFC to dispositional Self-Control that should be represented by the common variance of both Self-Control questionnaires. For the second-order factor model, loadings of Effortful Control and Trait Self-Control were set equal due to a high latent correlation of the first-order factors of.
Unstandardized loadings of first-order factors were set equal. Results of the analyses are displayed in Table 1. Thus, the second-order-factor model was used for all subsequent analyses. Standardized estimates, item parcels as manifest indicator variables see text for details. A Baseline model. B Moderation model, interaction term calculated with residual indicators following the procedure by Little et al. C Mediation Model. D Complete Mediation Model. In a next step, we tested moderation and mediation models summarized in Table 2. To test the moderation hypothesis, we followed the procedure by Little et al.
This model was compared to a more parsimonious model of complete mediation with fixing the direct path of NFC to Self-Control at 0. All three models are depicted in Fig 3 , panels B-D. Finally, we addressed the question of the actual power and robustness of the estimates obtained in the partial mediation model. With regard to power considerations, we used the semTools package for R to determine the actual power to detect the indirect effect of the partial mediation model following a method described by Satorra and Saris [ 64 ] where a model with the observed parameters is compared to a model with the parameter in question—here, the indirect effect—being fixed to zero.
- Topical Absorption of Dermatological Products?
- Middle East Quarterly (Spring 2009).
- US Army, Technical Manual, TM 55-1930-203-10, OPERATORS MANUAL FOR LIGHTER, AMPHIBIOUS, (LARC-LX), SELF-PROPE DIESEL, STEEL, 60-TON, 61 FT, DESIGN 2303, HULL 5 THRU 60, (NSN 1930-00-392-2981), 1975?
- About the Author?
In all cases, a power of 1 was achieved given our sample size. For each sample size, the partial mediation model was fitted to the data and the estimate of the indirect effect as well as the RMSEA was determined. Fig 4 gives the results. An integrative model of Self-Control identified different components and processes that precede successful Self-Control including, among others, the conflict between a current desire and higher-order goals as starting point as well as the motivation to control desire [ 31 ].
Cognition in Action Facility
Following this model of Self-Control components [ 31 ] and the theoretical conceptualizations of NFC and AO, AO was assumed as intervening variable moderating or mediating. Self-Control was assessed with two questionnaires that refer to Trait Self-Control but comprise slightly different aspects of Self-Control [ 56 , 58 ].
- Participate in Music Research.
- Recent publications.
- Cognition and action.
Therefore, we had to test a measurement model of Self-Control first. Both Self-Control measures correspond in that they measure dispositional Self-Control [ 56 , 58 ]. However, they are based on different theoretical backgrounds and hence focus on slightly different behaviors related to Self-Control [ 56 , 58 ].
Following this assumption of shared variance on a higher level and differences on a lower level, we assumed a common Trait Self-Control factor divided in two facets: a temperamental facet referring to Effortful Control with relations to basic attentional and cognitive processes [ 55 ] and a facet of Self-Control as dispositional tendency in everyday life [ 57 ].
Confirming this assumption, our results provide evidence for a hierarchical structure of two first-order factors that indicate both Self-Control facets and one higher-order factor of Self-Control that represents the large amount of shared variance.
To View More...
For the two questionnaires used in our study, this finding confirms previous research on the convergence of different Self-Control measures e. That research [ 54 , 67 ] has outlined that self-report measures of Self-Control share a large amount of variance attributed to dispositional tendencies toward Self-Control whereas performance tasks aim at specific control processes. The moderation model showed no significant interaction, indicating that the prediction of Self-Control by NFC was stable across different AO levels and rejecting the moderation hypothesis.
The partial mediation of this association through AO indicates that both traits are relevant for the prediction of Self-Control.
Imaging natural cognition in action.
Albeit the remaining direct effect of NFC being small, it is remarkable when keeping in mind that AO refers to processes very proximal to exerting Self-Control [ 31 , 42 , 43 ]. Hence, the direct predictive value of NFC was as small as expected but it still existed when accounting for the final recruitment of control resources represented by AO. This finding indicates that Self-Control depends not only on dispositions that refer to behavior very close to control processes AO but also on dispositions that are more broadly related to cognitive processes, such as NFC. Considering key components of Self-Control behavior [ 31 ], the current results suggest that different NFC levels have implications on resource allocation and on the engagement not only in general cognition but also in the implementation of control intentions.
The present results support the assumption that considering interindividual differences in personality adds to the understanding of the nature and preconditions of Self-Control [ 23 ]. The Self-Control measures used in this study refer to observable behavior that indicates successful dispositional Self-Control, which can be considered to be a proximate consequence of actually invested control effort. Hence, the higher predictive value of AO compared to NFC fits the conceptualization of AO that is theoretically rooted in research on action control [ 22 , 23 ].
From this perspective, it is even more remarkable that the current results provide evidence for NFC predicting Self-Control in part but not only through AO. Both dispositions are—at least in part—associated with different psychological processes. Hence, the partial mediation can be seen as indirect evidence for different psychological components and processes contributing to how individuals manage behavioral conflicts that demand Self-Control [ 31 ]. It indicates that the underlying mechanisms for the relation of NFC to Self-Control do not only refer to an increased tendency to recruit resources to control.
Instead, our results provide indirect evidence for relations of NFC with additional attentional and motivational processes relevant to Self-Control e. Referring to the marshmallow test and considering different processes that contribute to observable Self-Control [ 31 ], NFC may influence 1 to what extent a person is aware of the conflict between one marshmallow versus two, 2 to what extent one is motivated to solve the conflict with increased cognitive engagement, and 3 to whether one actually recruits control resources to manage waiting for the experimenter with strategies like cognitively focusing on the color of a marshmallow instead of the delicious taste.
For personality research in particular, our results suggest an association of higher NFC levels with increased tendencies to AO and thereby with rather flexible responses to situations that demand control. It suggests that the engagement aspect of NFC is important not only for thinking in general but also for dealing with situations that demand persistence, focusing on long-term goals, and Self-Control.
This finding is additional evidence for an association of NFC and the way individuals use their resources. The conceptual link of AO to affect regulation for an overview, see [ 23 ] further supports the conclusion of previous research that interindividual differences in NFC can have implications for affective adjustment e.
To measure Self-Control, we used two questionnaires that assessed self-perceived Self-Control generalized across different situations. This procedure took into account different theoretical approaches and referred to different behavioral aspects related to Self-Control. However, one could criticize that two self-report measures were used instead of laboratory performance tasks. One pragmatic argument for using self-reports is that they are less time-consuming when researchers aims at large sample sizes as basis for firm conclusions.
In contrast to performance tasks, they are likely to provide information representative for real-life behavior across different situations for a review on measurement characteristics, see [ 68 ]. Underscoring this notion and referring to the validity of Self-Control measures, questionnaires should be favored over single performance tasks when facing time or budget constraints [ 54 ].
Additionally, the current choice of Self-Control questionnaires was based on our aim to assess Trait Self-Control [ 54 , 67 ]. Following the complementary advantages of self-reports and performance tasks, the usage of questionnaires in the current study facilitated a large sample size and thereby well-founded results.