'وزن یک فکر: سرکوب افکار های مربوط به غذا در جوانان چاق و با وزن طبیعی

Objective The ironic processing theory by Wegner states that suppressing unwanted thoughts can be counterproductive because it leads to a rebound of these thoughts, eventually causing increased preoccupations. In line with this view, the present study examines the effects of suppressing food- and eating- related thoughts in obese and non-obese youngsters. Method Ninety-seven clinically obese and non-obese boys and girls, with different levels of dietary restraint, were asked to monitor their thoughts about food for 5 min, during three subsequent trial periods. Participants were instructed to suppress thoughts about food or to merely monitor them. The number of food-related thoughts was recorded across trials. Results No indication was found for an overall post-suppression rebound effect. However, when examining subgroups of high and low restrained eaters, results showed that the obese high restrained eaters displayed a pattern that was indicative of a rebound effect for food-related thoughts. None of the other groups shared these effects. Discussion The results corroborate the hypothesis that trying not to think about food and eating may be counterproductive, at least for a subgroup of clinically obese high restrained eaters. Thought suppression may be one of the factors contributing to acceleration and perpetuation of burdensome food-related thinking in clinically obese high restrained eaters.

The only way to get rid of a temptation is to yield to it. Resist it, and your soul grows sick with longing for the things it has forbidden to itself. Oscar Wilde (1891)—The Picture of Dorian Gray Trying not to think about a specific topic is a common way of dealing with unpleasant thoughts in everyday life. However, since the trail-blazing ‘white bear’ experiment (Wegner, Schneider, Carter, & White, 1987), clinical literature has acknowledged that deliberate attempts to suppress unwanted thoughts can lead to a paradoxical increase in the frequency of those thoughts (see Abramowitz, Tolin, & Street, 2001). Wegner et al. (1987) asked participants to suppress thoughts about a white bear, which ironically resulted in an escalation of those thoughts. In other words, thought suppression is not always an effective tactic of mental control, and may sometimes be counterproductive, leaving individuals with the very state of mind they had hoped to avoid (Wenzlaff & Wegner, 2000). As such, in Wegner's view thought suppression is a mechanism that can underlie, strengthen or maintain preoccupations. Some authors found evidence for an unwanted increase in target thoughts during suppression, which is known as the immediate enhancement effect (Wegner & Erber, 1992). The unwanted increase in thoughts or the recurrence of formerly successfully suppressed thoughts after suppression has ceased, is known as the ‘post-suppression rebound effect’ (Wegner, 1994). The paradoxical effects of thought suppression are not only found for uncommon thoughts (e.g. white bears), but also for dysfunctional preoccupations that crop up spontaneously in people with psychopathology. Indeed, thought suppression seems to play a key role in cognitive-behavioural models for a variety of emotional disorders, including depression (Wenzlaff & Bates, 1998), posttraumatic stress disorder (Shipherd & Beck, 1999) and in particular obsessive- compulsive disorder (Janeck and Calamari, 1999 and Smari et al., 1995). Despite the similarities between OCD and eating disorders (Anderluh, Tchanturia, Rabe-Hesketh, & Treasure, 2003), the interest in thought suppression in the field of eating-related problems is fairly recent and many researchers expressed the need for more research in this area (Herman and Polivy, 1993 and Wenzlaff and Wegner, 2000). Ward, Bulik, and Johnston (1996) were pioneers in hypothesizing that the suppression of thoughts about food, weight and body shape can lead to an increase in the frequency of those thoughts and hence contribute to the escalation and perpetuation of bulimic pathology. The common ground between the paradoxical effects resulting from thought suppression and the typical symptoms of eating problems is indeed remarkable. Therefore, this study investigates the role and contribution of thought suppression to eating-related pathology. More precisely, the consequences of suppressing thoughts about food and eating on the frequency of such cognitions will be examined both in obese and non-obese youngsters. An essential first step in order to implement thought suppression in everyday life is that there has to be a thought that one wishes to suppress. The experience of intrusive thoughts is quite common in the general population. Studies have shown that over 80% of the total population experiences occasional intrusions (Wells & Morrison, 1994). Yet, these thoughts are not all problematic. What distinguishes ‘normal’ intrusions from more pathological intrusions is that the latter appear to be more frequent, more intense, last longer, produce more discomfort and tend to evoke more resistance (Salkovskis & Harrison, 1984). Any one of these characteristics places an extra burden on individuals facing unwanted thoughts and increases attempts to control and suppress the unwanted thoughts (also see Purdon, 2004). Research has shown that food-, eating- and dieting-related thoughts are widespread in current western societies, even in children and adolescents (Kristensen, 2000, Nowak, 1998 and Shapiro et al., 1997). The contradiction between the thin ideal and the so-called toxic environment where palatable food and advertising are all around us, makes it hard not to focus on eating, at least once in a while (Brownell & Horgen, 2004). Although food thoughts need not be unwanted, e.g. when focussing on the gustatory pleasure of eating; focussing on the caloric threat of food or on past experiences of overeating or loss of control can induce thought suppression (Herman & Polivy, 1993). Research also indicates that hunger can induce thinking about food (Herman & Polivy, 1993) and increase the cognitive focus towards food cues (Mogg et al., 1998 and Placanica et al., 2002), especially in people who self-induce their hunger due to dietary restraint attitudes. Some groups appear to be more prone to develop obsessive food-related thinking than others. Eating-disordered individuals are the most obvious at-risk group. Their eating and weight preoccupations are a core characteristic and a diagnostic criterion for eating disorders (DSM-IV, APA, 1994). A vicious circle can arise in which a heightened focus on food and weight increases the risk for eating-problems and the eating-problems in their turn maintain and increase the focus on food and eating (Cooper and Todd, 1997 and Jones-Chesters et al., 1998). For related reasons, the obese population deserves special interest. Obese individuals tend to share a lot of the same stressors with eating-disordered patients, such as e.g. negative feelings about body weight and shape and failure experiences related to eating (Adami et al., 1994). Furthermore, obese youngsters show a similar bias in information processing, reflecting hypersensitivity for food cues (Braet & Crombez, 2003). Moreover, although direct studies on food- and eating-related thinking in obese individuals are scarce, Israel, Stolmaker, and Adrian (1985) found food-related thoughts to be more predominant among heavier individuals in a sample of undergraduates. This finding was replicated in an adolescent population, showing a strong positive association between Body Mass Index (BMI) and food-related thinking (Lynch, Eppers, & Sherrodd, 2004). Another similarity between eating-disordered and obese individuals is that both groups often show high levels of dietary restraint, denoting the cognitively mediated effort and intention to restrict food intake in order to control body weight or shape (Adami et al., 1994 and Herman and Mack, 1975). Previous research has found high restraint tendencies in obese children (Braet & Wydhooge, 2000) as well as in obese adults (Van Strien et al., 1986a and Van Strien et al., 1986b), although some obese are more intrinsically motivated to control their weight than others (Blokstra, Burns, & Seidell, 1999), illustrating the heterogeneity of this population. Individuals with dietary restraint attitudes appear to be another group at risk for food-related thinking, regardless of their actual current dieting behaviour. The association between dieting and food-related thinking was first brought to the fore by the early starvation studies by Franklin, Schiele, Brozek, and Keys (1948), which revealed that extreme dieting and associated weight loss can lead to a remarkable increase in the frequency of food-related thoughts. More recently, a study of undergraduate students by Mann and Ward (2001) showed the same findings in less severe circumstances. When individuals were prohibited to eat a particular food for 5 days, their food thoughts increased significantly. Ogden (2003) reached the same conclusions for dieters who self-imposed their diet. Furthermore, chronic dietary restraint was found to be more strongly associated with the selective processing of food cues than short-term food deprivation (Stewart & Samoluk, 1997). The question now is what role thought suppression might play in all this? As outlined above, it cannot be denied that food-related thinking is certainly prevalent in at-risk groups. Yet, the position of thought suppression can be twofold: first, it can strengthen preoccupations that potentially exist due to other factors (e.g. environmental factors, attentional bias, weight loss) and second, it can induce thoughts that are not yet obsessive in nature to become obsessive. Wenzlaff and Wegner, 1998 and Wenzlaff and Wegner, 2000 state that people who are highly motivated to control their cravings (e.g. smoking, drinking, eating) are often the very ones for whom inhibition seems to backfire, both in thoughts and behaviour. Such a strong motivation is very characteristic for both obese as well as non-obese restrained eaters, given their intent not to eat certain foods or to eat them less. As such, the denial of food becomes central to the individual's cognitive set and certain foods are seen as ‘forbidden’ (Ogden, 2003). Trying not to think about or avoid foods may be a logical consequence, that is, a suppression strategy. Indeed, thoughts about food and eating become symbols of danger for those wanting to control their eating and weight, and presumably make it harder to refrain from eating (Herman & Polivy, 1993). Given the ironic effects believed to be associated with thought suppression attempts, the presumed safety behaviour may actually fuel an exacerbation of food-thoughts, instead of reducing them and hence escalate food- related thinking. Some even state that the attempted suppression of food thoughts, rather than factors such as food-deprivation, actually produces the preoccupation with food (Herman & Polivy, 1993). Research on processes involved in the onset and maintenance of preoccupations is important, especially because it points out that preoccupations with dieting- and weight-related thoughts deplete cognitive resources otherwise available for non-dieting tasks (Jones & Rogers, 2003), predict compulsive eating (Ghiz & Chrisler, 1995) and play a meaningful role in causing and maintaining binge eating disorder (Adami et al., 1995 and Adami et al., 1998). To explain the paradoxical effects associated with thought suppression, Wegner, 1992 and Wegner, 1997 and Wegner and Zanakos (1994) state that the intention to suppress a thought initiates two complementary cognitive processes. The first process, the ‘intentional operating process’ initiates a conscious search for distracters, that is, for any unrelated mental material other than the unwanted thought. As such, this controlled search for distracters is the mental operating process that bears out the actual suppression. It is an effortful, controlled process that requires a great deal of mental capacity. Simultaneously, a second process called the ‘ironic monitoring process’ is set in motion. This unconscious process is constantly on the lookout for the unwanted thought. Its task is to signal the individual if the unwanted thought should appear. As such, it alerts the operating process to renew distraction when conscious awareness of the unwanted thought becomes imminent (Wenzlaff & Wegner, 2000). It is a less effortful, automatic mechanism demanding less mental capacity. When mental capacity is inadequate (e.g. due to cognitive loads like stress or time pressure) or when suppression attempts are voluntarily terminated, the intentional operating process is restricted or halted and no longer provides distracting thoughts. Yet, the less effortful ironic monitoring process keeps on scanning for the unwanted thought, brings it to mind, and leaves the individual with a failed attempt to suppress a thought. This ironic effect increases mental load and triggers the whole mechanism once more. Wegner (1994) calls this the self-loading ironic system. According to Wegner and Zanakos (1994), the rebound of a thought, following its suppression, can be so intense that it creates preoccupations. Further studies have indicated that the emotional charge of the suppressed thoughts also plays a part in the suppression effects (Wenzlaff & Wegner, 2000). Neutral thoughts or thoughts not relevant to current concerns are probably easier to suppress and less likely to trigger ironic effects (Davies and Clark, 1998, McNally and Ricciardi, 1996 and Roemer and Borkovec, 1994), although some controversy about this matter remains (see Abramowitz et al., 2001). To clarify the role of thought suppression in promoting and maintaining food-related thinking in obese and non-obese adolescents, the present study addresses the following research questions. First, in line with the ironic processing theory by Wegner (1994), our study expects to find an overall rebound effect for food- and eating-related thoughts, following experimentally induced thought suppression. Given that it is more difficult to suppress thoughts that are emotionally charged (Davies & Clark, 1998) or that are relevant to current concerns (McNally & Ricciardi, 1996), and granted that material that is harder to suppress may yield a stronger rebound effect (Roemer & Borkovec, 1994), we further hypothesize that the effects of thought suppression will be the strongest for those who are more intrinsically motivated to control their weight, in case the high restrained eaters. We expect these effects for both the obese and non-obese subjects.

Descriptive statistics Comparing the obese and normal-weight samples, revealed a significant difference between the two weight-groups for BMI (t (79)=15.06, p<0.001), adjusted BMI (t (80)=15.93, p<0.001), dietary restraint score (t (82)=2.17, p=.03), age (t (94)=−4.09, p<0.001) and feelings of hunger at baseline (t (85)=−2.74, p=0.01). There were no gender differences between the two weight-groups (χ2=0.04, p=0.84). The first two differences were expected based on the participants' weight status. The fact that age and the baseline rating of hunger were higher in the normal-weights when compared to the obese group was not in line with expectations. In contrast, the finding that the obese obtained higher average dietary restraint scores than the normal-weights corresponds to previous research findings in children, adolescents and adults (Braet and Van Strien, 1997 and Van Strien et al., 1986a). When looking more closely at the distribution of dietary restraint scores across restraint and weight status, no further group differences were found. An ANOVA (F (3,93)=69.31, p<0.001) and Turkey HSD post hoc analyses revealed no significant differences between the obese low restrainers and the normal-weight low restrainers on dietary restraint scores (p=0.34). The same was true when comparing the obese high restrainers with the normal-weight high restrainers (p=0.71). As expected and indicated, the restraint scores of the obese high restrainers were significantly higher than the restraint scores of the obese low restrainers (p<0.001). This is also the case when comparing the normal-weight high and low restrainers on mean restraint scores (p<0.001). Table 1 provides an overview of the participants' characteristics per group.