روشن ساختن نقش تکانشگری در خویشتن داری غذایی: روش مدل سازی معادلات ساختاری

This study was designed to clarify the relationship between the four dimensions of impulsivity in Whiteside and Lynam’s (2001) model and the two aspects of dietary restraint (Concern for Dieting and Weight Fluctuation) in a non-clinical sample. Data were collected from a volunteer community sample (N = 216) of women who responded to two self-report instruments related to impulsivity (UPPS Impulsive Behavior Scale, Whiteside & Lynam, 2001) and dietary restraint (Restraint Scale, Polivy, Herman, & Warsh, 1978). A structural equation model was tested. The model provided a good fit to the data (χ2/df = 1.64, p < .0001, RMSEA = 0.054, 90% CI = 0.050–0.058, p-value for test of close fit (RMSEA > 0.05) = .035) and revealed that Concern for Dieting was positively related to Urgency (standardized β = 0.25, p < .01). Weight Fluctuation was related to lack of Perseverance (standardized β = 0.25, p < .01) and tended to relate to Sensation Seeking (standardized β = 0.17, p = .056). Urgency and lack of Perseverance were found to play a significant role in predicting Concern for Dieting and Weight Fluctuation.

Impulsivity is a multidimensional concept that has been conceptualized in various ways and that incorporates “actions that are poorly conceived, prematurely expressed, unduly risky, or inappropriate to the situation and that often result in undesirable outcomes” (Evenden, 1999, p. 348). Impulsivity also involves a tendency to seek sensations and rewards (Evenden, 1999). A number of studies have shown that different aspects of impulsivity are related to eating disorders in clinical (e.g., Claes, Vandereycken, & Vertommen, 2005) and in non-clinical populations (e.g., Lyke & Spinella, 2004). More specifically, it has been found that motor impulsivity (indicating proneness to imprudent actions), cognitive/attention impulsivity (indicating inability to maintain focused attention) and heightened reward sensitivity are associated with problematic eating behaviors, such as overeating, dieting and weight fluctuation. Loxton and Dawe (2001) revealed that greater sensitivity to both reward and punishment predicts dysfunctional eating in 16- to 18-year-old girls. Furthermore, Guerrieri, Nederkoorn, and Jansen (2007) showed that high levels of impulsivity is associated with higher food intake in a population of undergraduates. Research on impulsivity and dietary restraint and eating disorders has relied on many different self-report scales that measure a wide range of impulsive behaviors. Recently, Whiteside and Lynam (2001) clarified the various conceptions of impulsivity by means of factor analyses conducted on 10 commonly used impulsivity measures; these measures revealed four components associated with impulsive behaviors. The four components, which are the basis for the UPPS Impulsive Behavior Scale, are: (1) Urgency, defined as “the tendency to experience strong impulses, frequently under conditions of negative affect”; (2) lack of Premeditation, defined as “the tendency not to think and reflect on the consequences of an act before engaging in it”; (3) lack of Perseverance, defined as “the inability to remain focused on a task that may be boring or difficult”; (4) Sensation Seeking, defined as “the tendency to enjoy and pursue activities that are exciting, and openness for new experiences”. Studies using the UPPS Impulsive Behavior Scale conducted on non-clinical samples (e.g., Anestis et al., 2007 and Fischer et al., 2003) revealed that bulimic symptoms (loss of control over eating, vomiting) were positively related to Urgency, but not to the other facets of the scale. However, two other studies, which used the revised NEO-Personality Inventory (Costa & McCrae, 1992) in both clinical (Claes et al., 2005) and non-clinical populations (Miller, Flory, Lynam, & Leukefeld, 2003), revealed that bulimic symptoms are positively related to Urgency, lack of Premeditation and lack of Perseverance, although the latter relations were weaker than the relation with Urgency. In the Claes et al. study (2005), patients with bulimia showed more Urgency, more Sensation Seeking, less Perseverance and less Premeditation than patients with restrictive anorexia nervosa, while patients with bulimia and binging/purging anorexia nervosa did not differ significantly from each other with respect to impulsivity-related traits, except for Urgency, where patients with bulimia scored higher. Previous studies of non-clinical populations focused on the relationship between impulsivity and bulimic symptoms, but the relationship between dietary restraint and impulsivity has received a fair amount of attention. Dietary restraint defines a type of eating behavior that is governed by cognitive processes (Concern for Dieting) rather than by physiological mechanisms such as hunger and satiety. Dietary restraint is used to achieve or maintain a desired body weight; however, high levels of dietary restraint are not associated with successful weight control because people with high levels of dietary restraint are also prone to episodes of overeating (Heatherton, Herman, Polivy, King, & McGree, 1988). Over time, a succession of restraint and overeating episodes lead to weight fluctuation and even eating disorders. Thus, a clearer understanding of the association between dietary restraint and different facets of impulsivity is important for a better understanding of the development of eating disorders. Dietary restraint can be assessed by the Restraint Scale and its Concern for Dieting and Weight Fluctuation subscales. In this context, the objective of the present study was to explore how the four dimensions of Whiteside and Lynam’s model of impulsivity are linked with dietary restraint in a non-clinical sample. It was hypothesized that lack of Perseverance and high Sensation Seeking would be associated with Concern for Dieting, and that Urgency and lack of Premeditation would be associated with Weight Fluctuation. As proposed by Dawe and Loxton (2004), reward sensitivity may contribute to greater sensitivity and attention towards food-related stimuli (e.g., TV commercials) and therefore be associated with a desire to eat and Concern for Dieting. Lack of Perseverance may be associated with Concern for Dieting via a difficulty controlling thoughts related to food and body shape (body dissatisfaction). In this context, Urgency and lack of Premeditation may contribute to the inability to resist eating and to loss of control over eating, and therefore contribute to Weight Fluctuation. To explore the associations between impulsivity and the two aspects of dietary restraint, structural equation modeling (SEM) was applied. The main advantage of SEM in this context is its capacity to simultaneously estimate all effects of the four impulsivity factors on the two dietary restraint dimensions. Moreover, SEM allows one to adjust for measurement errors, thereby providing unbiased estimates of the potential associations between the impulsivity and dietary restraint constructs. These aims cannot be attained with standard regression analysis.

The aim of this study was to identify which facets of impulsivity are associated with dietary restraint in a non-clinical sample of females. The French versions of the Restraint Scale (Lluch, 1995) and the UPPS Impulsive Behavior Scale (Van der Linden et al., 2006) were administered to 216 women from the general population. The main results of the study may be summarized as follows. First, the SEM model revealed that more Concern for Dieting was reported by individuals with high Urgency. Secondly, more Weight Fluctuation was reported by individuals with low levels of Perseverance and high levels of Sensation Seeking. Our results showed a positive association between Concern for Dieting and Urgency. Bechara and Van der Linden (2005) tentatively proposed that Urgency may be related to the inability to deliberately suppress dominant or automatic responses, especially in conditions of intense emotions. In this context, the positive association between Concern for Dieting and Urgency suggests that overeating, which may follow dietary restraint, is an automatic response used to relieve negative affect in the short term; it may increase the likelihood of excessive food intake. Some authors call this phenomenon “emotional eating”, in other words, overeating in response to negative emotions (e.g., anxiety, sadness, and anger). Our results are also consistent with those of Bekker, van de Meerendonk, and Mollerus (2004), who found a positive link between impulsivity and emotional eating. This inability to prevent oneself from over eating may have detrimental immediate (e.g., self-depreciation, depressed mood, abdominal pain) and long-term consequences (e.g., weight gain), and therefore may lead to concern for dieting. Our results showed a positive association between Weight Fluctuation and lack of Perseverance. This association is in line with Lyke and Spinella (2004) study, which revealed that a person’s sense of loss of control over eating, was positively correlated with both attentional and motor impulsivity. According to Bechara and Van der Linden (2005), lack of Perseverance should be related to resistance to proactive interference, which refers to the inability to inhibit irrelevant thoughts or memories. In the case of dietary restraint, individuals may have difficulties controlling their thoughts of food or thoughts concerning their shape and weight. In fact, it seems that high-impulsive individuals report more eating, weight and shape concerns than do low-impulsive individuals (Guerrieri et al., 2007). This phenomenon may lead to overeating because thoughts of food increase the desire to eat (Johnston, Bulik, & Anstiss, 1999). In addition, Engel et al. (2006) showed that attentional avoidance of body and shape concerns, a dysfunctional cognitive strategy, increases such concerns. Alternatively, eating may be used to escape negative thoughts (e.g., general negative thoughts or self-depreciating thoughts following out-of-control eating). This phenomenon is probably reinforced by the temporary alleviation of the negative thoughts and emotions and it may well increase the likelihood of excessive food intake (see the “Escape Theory of Binge Eating”, Heatherton & Baumeister, 1991). We also found a weak positive link between Weight Fluctuation and the third facet of impulsivity, Sensation Seeking. This result is comparable to those obtained by Jansen, Klaver, Merckelbach, and Van den Hout (1989), who reported that restrained eaters scored higher on Sensation Seeking scales. Sensation Seeking may be related to a tendency on the part of restrained eaters to exaggerate the impact of rewards (Bechara, Dolan, & Hindes, 2002). This tendency to seek rewards (food) may lead to weight gain. We must also discuss the absence of any association between dietary restraint and lack of Premeditation, which is consistent with the results of Fischer et al. (2003) in a non-clinical population; they showed that UPPS-Urgency was positively related to bulimic symptoms, whereas UPPS-lack of Premeditation was not. Bechara and Van der Linden (2005) hypothesized that lack of Premeditation is related to the inability to take into account the positive or negative consequences of a decision on the basis of the emotional responses associated with it (see the somatic marker hypothesis, Damasio, 1994). The absence of any relationship between dietary restraint and lack of Premeditation suggests that dietary restraint arises from an inability to inhibit a problematic behavior and thoughts, rather than from failure to consider the negative consequences of the dietary restriction. Nevertheless, we cannot rule out the possibility that lack of Premeditation contributes to more severe disordered eating and is implicated at various stages of the binge-purge cycle. In fact, persons with problematic eating are characterized by repeated loss of control over their eating in spite of the negative consequences of their behavior (e.g., negative affect, physical impairments, isolation). This study confirmed and extended previous results regarding the relationship between impulsivity and dysfunctional eating. Indeed, the results of the current study indicate that all dimensions of impulsivity except lack of Premeditation are positively related to dietary restraint. This study allows us to focus on several interesting new perspectives that will contribute to a better understanding of dysfunctional eating. It may be used to help predict and prevent restrained eating. Indeed, Carter, Stewart, Dunn, and Fairburn (1997) have demonstrated that an eating-disorder prevention program designed to reduce dietary restraint did more harm than good. In their view, the prevention of eating disorders can only be effective if it focuses on identified groups and takes into account the characteristics of the high-risk group. Although the findings of this study provide important information concerning the relationship between impulsivity and dysfunctional eating, several limitations are worth noting. First, the design of this study is cross-sectional, which precludes the possibility of establishing causal links between impulsivity and dietary restraint constructs. Indeed, it must be noted that, as is commonly the case for any SEM, there is an infinite number of equivalent models that will produce the exact same statistical fit but completely change the theoretical perspective. Consequently, we cannot infer causality from these data. However, the use of SEM allows one to test complex theories and control for differential item reliability, neither of which is possible with classical analyses such as multiple regression or analysis of variance. Second, the degree to which impulsivity predicts dietary restraint is limited. Indeed, only 7% and 9% of the variance in Concern for Dieting and Weight Fluctuation, respectively, was explained by impulsivity. This is reflected in the high residual correlation between the dieting factors (r = 0.65). Clearly, there is much information that is common to both dieting factors but independent of impulsivity. In other words, other factors besides impulsivity probably underpin dietary restraint. It is possible that variables not measured here, such as preoccupying cognitions about dieting, body and weight, could account for dietary restraint. Another limitation of the study is the absence of a clinical comparison group and the fact that it was limited to women with a low BMI, which limits the generalizability of our findings. This low BMI could be, at least partly, due to an underestimation of weight ( Jacobson & DeBock, 2001), even though the underestimation of weight has been disproportionately associated with those who are overweight/obese ( Larsen, Ouwens, Engels, Eisinga, & Van Strien, 2008). If this underestimation exists, it does not constitute a major problem to the extent that even an underestimation of 3 kg (the mean error of self-reported weight observed in Larsen et al., 2008) would not change the normality of the mean BMI of our sample. The self-report bias may also have influenced the respondents’ answers to the questionnaires. However, the use of SEM may have controlled for these measurement errors to some extent. Further studies should expand the scope of the proposed model to compare Whiteside and Lynam’s four impulsivity dimensions in both clinical (bulimia and binge eating disorders) and non-clinical control samples. The next step may be to use the model in a group of both males and females. Another promising direction for expanding the model could be a longitudinal examination of the relationship between impulsivity and dysfunctional eating. This extension would allow stronger inferences about causality. Lastly, further studies are required to explore the assumptions arising from our results. More specifically, it would be interesting to explore the cognitive processes underlying the four facets of impulsivity by using cognitive tasks specifically designed to explore these processes (see Bechara & Van der Linden, 2005, for suggestions) and thereby circumvent self-report biases.