Individuals with restrained eating employ restrictive diets to control their weight [26]. These diets have been associated with increased levels of depression, stress, abnormal attitudes toward weight and eating, and low self-esteem [20]. Notably, restrained eating reflects individuals' cognitive and behavioral efforts to manage their weight [39] and does not necessarily lead to lower BMI [16, 27]. To maintain their restrictive diets, individuals with restrained eating often rely on strict and rigid rules regarding their food choices and meal timings [26]. Adhering to such rules requires the active use of inhibitory control, namely, the ability to suppress or delay unwanted, irrelevant, or distracting behaviors, thoughts, or emotions [3, 9, 12, 21, 25, 26]. The connection between food and inhibition in restrained eaters has been demonstrated in several studies, indicating that food-related cues trigger inhibition in this population [12, 43]. Simultaneously, food-related cues also trigger explicit negative attitudes in these individuals, such that restrained eaters exhibit heightened evaluations of the negative aspects of high-calorie foods [32, 33]. Emerging evidence suggests a robust interconnection between these seemingly disparate phenomena, namely inhibitory control and food-related attitudes. For example, researchers have demonstrated that intense execution of inhibition toward a stimulus can result in a less positive attitude toward that stimulus [7, 11, 44]. Hence, the goal of the current study is to investigate whether disassociating food-related cues and inhibition, using a computerized task, can reduce the negative association of food cues in restrained eaters.

The association between food and inhibition has been the subject of numerous investigations. The term food-inhibition association [23] refers to the effect of exposure to food stimuli on inhibitory abilities. In the general population, exposure to food stimuli, generally leads to reduced inhibition (i.e., more food consumption) [12, 13, 29] and increased food cravings [31] when compared to exposure to neutral non-food stimuli. Some researchers have discussed this finding in light of the evolutionary importance of eating, which led to researchers suggesting that there are specific neuronal and behavioral mechanisms meant to ensure appropriate food consumption [8, 37, 40]. Interestingly, individuals with restrained eating have shown an increased level of inhibition when presented with high-calorie foods, compared to their baseline inhibition [12], a finding that has also been replicated in patients with anorexia nervosa [49]. This intriguing effect, which can be viewed as a 'reversal' of evolutionary processes, has been suggested to play a crucial role in the development and maintenance of disordered eating [17]. Specifically, it has been proposed that years of practicing inhibition in response to food stimuli can lead to the formation of a strong food-inhibition association in individuals with disordered eating [50]. This food-inhibition association can significantly influence the attitudes toward food in these individuals and may lead to the automatic inhibition of eating behaviors in response to food stimuli, thereby maintaining restrained eating behaviors.

The relationship between inhibition and attitudes/emotions is a complex and dynamic one [38]. Repeated inhibition targeted at a specific stimulus can exert profound effects on attitudes and emotions associated with said stimulus [11, 44]. This phenomenon can be explained through the Behavior Stimulus Interaction (BSI) model [44]. According to this framework, when a stimulus elicits an approach reaction, but environmental cues indicate that approaching the stimulus is unfavorable (and, thus, the approach behavior should be inhibited), a dynamic interplay emerges between the stimulus and behavioral inhibition, resulting in the modification of attitudes toward the stimulus. Veling et al. [44] demonstrated that the repetitive activation of inhibition toward positive stimuli leads to their devaluation. For example, imagine an individual who loves dogs and then gets bitten by one. According to the BSI model, if that individual will then consistently suppress or inhibit their responses to this particular stimulus (dogs), then over time, the individual will become increasingly sensitized to the stimulus, and develop negative attitudes and emotional reactions toward the stimulus. Consistent with this model, research shows that when food, which commonly serves as a positive stimulus that evokes approach behavior, is linked to inhibition, individuals will rate those foods as less liked [7, 45]. For example, Chen et al. [6] showed that associating specific foods with inhibition (by asking participants not to respond to trials of a cognitive task containing these foods) leads to the devaluation of these food stimuli (see also: Houben [19]). Hence, in the case of individuals with restrained eating, the persistent and repetitive inhibition toward food may potentially amplify pre-existing negative attitudes toward food, perpetuating a vicious cycle of both negative attitudes toward food and disordered eating behaviors.

Several therapeutic interventions have focused on the association between food and inhibition. Specifically, in recent years, researchers have explored the use of cognitive tasks to manipulate the food-inhibition relationship to reduce clinical symptoms [5, 28, 41]. One common approach to modulate this relationship is by utilizing a cognitive task that requires inhibitory control, such as the stop-signal task [24, 47]. While originally designed to assess inhibitory control, researchers have employed this task, and similar ones, to create a new 'learned' reflexes by associating target stimuli with inhibition [46]. For instance, Hochman et al. [18] manipulated the proportion of specific trials that necessitated inhibition. This was done by ensuring that a high proportion of the conditioned stimulus (e.g., a green circle) was consistently followed by a stop-signal, which indicated to the participant the need for inhibition. Meanwhile, the unconditioned stimulus (e.g., a blue rectangle) was rarely followed by a stop-signal. Through this approach, researchers were able to condition levels of automatic inhibition triggered by the conditioned stimulus (see also: Manasse et al. [28]). In the context of eating behaviors, researchers have utilized similar tasks to manipulate the automatic association between food cues and inhibition. In studies involving healthy controls, training tasks were employed to condition different types of food with either inhibition or response, revealing that participants subsequently exhibited a preference for food associated with response over food associated with inhibition [1, 30, 45]. In addition, overweight and obese participants who underwent training associating high-calorie snacks with inhibition displayed a decrease in their liking of these snacks and these participants also lost more weight over time [22]. Thus, taken together with the above research, it is evident that the food-inhibition relationship is not only central to disordered eating but can also be manipulated using computerized tasks.

Until recently, researchers focused on creating a food-inhibition association to examine how such an association can facilitate weight loss. The question then arises whether the food-inhibition association can also be manipulated to decrease the inhibitory response to food stimuli, which would help individuals with restrained eating return to healthier approach attitudes toward food. Weinbach et al. [48] employed a similar training approach with individuals with restrained eating, this time focusing on establishing an association between food and response/disinhibition. Participants completed a modified version of the stop-signal task, wherein neutral images (e.g., table, chair) were frequently followed by a stop cue, while food-related images were never accompanied by such a stop cue. The results of this study indicated that following this training procedure, individuals with restrained eating were more likely to increase their food consumption in an unrelated bogus taste test. Interestingly, individuals with restrained eating who underwent the food-response training also reported higher food-related anxiety after the task, compared with their pre-training food-related anxiety levels. This study also included another group of restrained eaters who underwent similar inhibitory training, but without a specific association between food and response/inhibition, as stop-signals appeared equally in both food and non-food trials. This group, which was initially a control group, exhibited more positive attitudes toward food and even a reduction in their food-related anxiety [48]. The researchers suggested that this may be because this condition allowed for a more flexible approach to food, enabling the participants to alternate between response inhibition and response execution when exposed to food stimuli. However, it is important to note that a food-inhibition group was not included in that study, which limits our ability to draw definite conclusions about the impact of these training procedures on attitudes towards food, especially in the case of individuals with restrained eating.

The literature reviewed above provides evidence that different levels of the food-inhibition association can lead to the devaluation of food cues and, perhaps, can even increase food-related anxiety. Consequently, if we want to prevent the devaluation of food, and perhaps even to improve attitudes toward food, it is imperative to explore strategies that can effectively counter food-related anxiety and foster positive attitudes toward food. Previous studies suggest that utilizing a training procedure aimed at enhancing inhibitory flexibility can lead to a positive emotional evaluation of food, which can persist even after eating, among individuals with restrained eating [48]. Such training seeks to establish a balance between inhibiting and responding to food cues. Therefore, the goal of the current study was to investigate whether an inhibitory task designed to enhance flexible inhibition toward food would result in more positive attitudes toward food, especially compared to both food-response and food-inhibition training groups.

To achieve this aim, the experiment included three groups, each performing a different version of the food stop-signal task: The first group, referred to as the flexible response/inhibition group, completed the task in which food stimuli were equally associated with both responses and inhibition. The second and third groups, named the food-response and food-inhibition groups, completed the task in which food stimuli were consistently associated with either response or inhibition, respectively. Notably, the task consisted of only one brief training session. Pre- and post-training implicit association tests were used to measure implicit attitudes toward food. To ensure that the effect of the training persists even after food consumption, participants were also asked to eat as part of a seemingly unrelated Bogus taste test. Based on the literature reviewed earlier, we hypothesized that the flexible response/inhibition group would exhibit more positive attitudes towards high-calorie foods after completing the food stop-signal task, compared to both other groups. Furthermore, we anticipated that this improvement would persist even after eating. Additionally, if the brief (single session) training affects eating, we expected that food consumption would be highest in the food-response group, lowest in the food-inhibition group, and intermediate in the flexible response/inhibition group.