Introduction

Orthorexia nervosa (ON) is a theorized condition in some people who are pathologically preoccupied with healthy nutrition and who strictly follow a restrictive diet believed to be healthy and pure [1]. Beyond simply being health fanatics, individuals with ON experience a host of negative feelings and impairments that include intense feelings of guilt and shame when violating the rigid rules of their diet, psychosocial impairments whereby their ON obsessions and compulsions negatively impact their personal relationships and their vocational or academic performance, and physical impairments, such as malnutrition and associated health complications that result from their overly restrictive diet [2,3,4,5,6,7]. Although ON is not presently recognized as a disorder in either the International Classification of Diseases, Eleventh Revision (ICD-11) [8] or the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [9], a growing number of patients have been seeking help for ON-related impairments [4, 5, 10].

Research on ON reveals that the condition shares certain characteristics with both obsessive–compulsive disorder (OCD) and anorexia nervosa (AN). Regarding the former, ON symptoms are associated with greater obsessive–compulsive tendencies [11,12,13,14,15], consistent with research showing that individuals with ON experience obsessions about healthy nutrition, in addition to compulsions that include the need to eat in a ritualized way [15, 16] and the need to complete a rigid exercise regime [17,18,19]. Regarding the latter, ON symptoms are associated with greater disordered eating behaviors that characterize people with AN [12,13,14, 20, 21], and both AN and ON are associated with greater levels of anxiety, perfectionism, and excessive control over one’s environment including their diet [12, 16].

Another potentially shared characteristic among these disorders could be deficits in inhibitory control, or the inability to suppress an action or thought. Although people with ON would conceivably require a greater amount of self-control to follow their very restrictive diet, several studies reveal that restrained eating is associated with deficits in inhibitory control [22,23,24,25]. In the only published study investigating the potential association between ON and inhibitory control, Koven and Senbonmatsu [15] found that an ON group did not differ from the control group on the Stroop Task, which is a well-known assessment of inhibitory control, or response inhibition. Regarding the other two disorders, several studies have demonstrated deficits in inhibitory control through impaired performance on the Stroop Task, Go/No-Go Task, and Arrow Flanker Task in individuals with OCD [26,27,28,29,30,31,32] and in individuals with AN [33,34,35].

The current study seeks to extend the work of Koven and Senbonmatsu [15], but using three different measures of inhibitory control: the Go/No-Go Task, Arrow Flanker Task, and Stroop Task. Based on the commonalities between ON and both OCD and AN, along with previous research demonstrating impaired response inhibition in both OCD and AN, we may hypothesize that greater ON symptomatology would be associated with worse performance on these tasks, suggesting that a similar neuropsychological profile underlies all three conditions. In contrast, it is possible that we may find no such association, which would be consistent with Koven and Senbonmatsu’s finding that Stroop performance did not differ between the ON and control groups, suggesting that despite some overlap between ON and AN, ON may indeed be a unique condition, at least with respect to inhibitory control.

Methods

Participants

The participants were 63 undergraduate students (50 women, 13 men) enrolled in undergraduate psychology courses at Texas State University. They ranged in age from 18 to 34 years (M = 21.29, SD = 2.50), and most of them self-reported being White (43%), Hispanic/Latinx (35%), or Black (13%). All participants provided informed consent, and were entered in a drawing for a chance to win a $25 gift card.

Procedure and measures

Participants first completed the Eating Habits Questionnaire (EHQ) [13], a 21-item inventory for which subjects use a 4-point Likert scale to indicate how true each item is for them. Eight items assess healthy eating behaviors (e.g., “My eating habits are superior to others”), nine assess impairments that are associated with their commitment to healthy eating (e.g., “My healthy eating causes significant stress in my relationships”), and four assess feelings associated with healthy eating (e.g., “I feel in control when I eat healthily”) [36]. The reliability and validity of the EHQ has been established by the original authors of the inventory and by other researchers [36, 37]. For the current study, the EHQ demonstrated good internal consistency with Cronbach’s alpha and McDonald’s omega each equaling 0.89, and the participant scores ranged from 25 to 70 (M = 41.27, SD = 10.05).

Next, participants completed computerized versions of the Go/No-Go Task [38], Arrow Flanker Task [39], and Stroop Task [40] using the Inquisit software (millisecond.com). For the Go/No-Go Task, on each trial, a white rectangle appears on the screen and then turns either green or blue. Participants are instructed to press the spacebar key as quickly as possible when the rectangle turns green (i.e., Go trial), and to not respond if the rectangle turns blue (i.e., No-Go trial). The variables for this task are percent error on Go trials (i.e., omission errors whereby participants fail to respond), percent error on No-Go trials (i.e., commission errors whereby participants incorrectly respond), and response time (RT) on Go trials. Deficits in inhibitory control are associated with greater error on No-Go trials.

For the Arrow Flanker Task, on each trial, five arrows are presented inside a box. Participants are instructed to press the ‘Q’ key (on the left portion of the keyboard) if the center arrow points left, or to press the ‘P’ key (on the right portion of the keyboard) if the center arrow points right. Sometimes, all five arrows point in the same direction (i.e., congruent trial), and sometimes the center arrow points in the opposite direction than the rest (i.e., incongruent trial). The variables for this task are percent error and RT on congruent trials, and percent error and RT on incongruent trials. Deficits in inhibitory control are associated with greater error and greater RT on incongruent trials.

For the Stroop Task, on each trial, a word in a certain color font is presented on the screen. Participants are instructed to indicate the font color, ignoring what the word means, by pressing one of the following keys: ‘D’ for red, ‘F’ for green, ‘J’ for blue, or ‘K’ for black (this instruction appears on the screen on every trial). Sometimes the word is a color word in the same font, such as the word ‘RED’ typed in red font (i.e., congruent trial), sometimes the word is a color word in a different font, such as the word ‘GREEN’ typed in red font (i.e., incongruent trial), and sometimes the word is a non-color word, such as the word ‘TABLE’ typed in red font (i.e., control trial). The variables for this task are percent error and RT on congruent trials, percent error and RT on incongruent trials, and percent error and RT on control trials. As with the Arrow Flanker Task, deficits in inhibitory control are associated with greater error and greater RT on incongruent trials.

Results

Pearson correlation analyses were conducted to evaluate the relationships between EHQ scores and each variable from the three tasks assessing inhibitory control. As shown in Table 1, none of these relationships were significant.

Table 1 Correlations between EHQ scores and inhibitory control measures
Full size table

Discussion

The results revealed that ON symptomatology is not associated with deficits in inhibitory control as measured with three commonly used neuropsychological tasks. These results are consistent with those of Koven and Senbonmatsu [15], who found no differences between ON and control groups on the Stroop Task. The highly restrictive diet of people with ON undoubtedly involves restrained eating, and regarding the research question for the current study, past research shows that restrained eating is associated with deficits in inhibitory control [22,23,24,25], which leads to the question concerning what makes ON different and possibly not susceptible to such deficits. Some researchers point out the necessity to differentiate unsuccessful restrained eaters who partake in counterproductive disinhibited overeating (that leads to attempts at more extreme restrained eating that inevitably fail) from successful restrained eaters who exhibit the self-control needed to follow a restrictive diet [24]. Perhaps, people with ON are successful restrained eaters and their self-control with food extends to adequate inhibitory control in general, which is consistent with the lack of deficits found in the current study.

ON also shares commonalities with AN, whereby both ON and AN are associated with greater disordered eating behaviors [12,13,14, 19,20,21], as well as greater levels of anxiety, perfectionism, and excessive control over one’s environment [12, 16]. These commonalities comprise the primary reason underlying the hypothesis that ON would be associated with deficits in inhibitory control, because the past research reveals that people with AN exhibit such deficits through impaired performance on the Stroop Task, Go/No-Go Task, and Arrow Flanker Task [33,34,35]. However, ON and AN have been shown to differ with respect to other aspects of executive function too. Namely, whereas people with AN exhibit deficits in cognitive flexibility through impaired performance on the Wisconsin Card Sorting Task and Trail Making Task [41,42,43,44,45], ON and control groups have not differed on such tests [15, 46]. Thus, despite some overlap between ON and AN, ON appears to be a unique condition, at least with respect to inhibitory control and cognitive flexibility.

This conclusion should be interpreted with some caution, though, as this study has some potential limitations. Most importantly, this study was limited to a non-clinical sample of college students, such that the higher EHQ scores may have only been indicative of ON risk rather than true pathological impairment. As indicated previously, though, ON is not presently recognized as a disorder in either the ICD-11 [8] or the DSM-5 [9], which inherently limits the recruitment of individuals with an ON diagnosis. In future studies, researchers should recruit participants from community settings, such as health centers and eating disorder clinics, as well as through Instagram advertisements that specifically target people who are committed to healthy and clean eating. Such recruitment measures should serve to increase the range of clinical ON symptoms. In these future studies, researchers should also control for comorbidity with AN and OCD, as well as variables, such as intelligence and perfectionism that may impact performance on inhibitory control tasks. By expanding recruitment efforts, using stricter inclusion criteria, and controlling for potentially confounding variables, we may better determine whether ON is associated with impaired inhibitory control or whether ON is unique from AN in this regard.

What is already known on this subject?

Multiple studies reveal that both AN and OCD are associated with deficits in inhibitory control, but only one study assessed possible deficits with ON, finding that an ON-symptoms group did not significantly differ from a control group on the Stroop Task.

What does this study add?

This study explored potential associations between ON symptomatology and three measures of inhibitory control: Go/No-Go Task, Arrow Flanker Task, and Stroop Task. No significant relationships were found, suggesting that ON is not associated with deficits or other differences in inhibitory control.