Jayden Goodwin and Dr. Michael Larson, Psychology Department

Introduction

Perfectionism, or the pursuit of error-free performance, is often associated with shame, guilt, failure, and low self-esteem (Kilbert, Langhinrichsen-Rohling, & Saito, 2005; Stahl, Acharki, Kresimon, Völler, & Gibbons, 2015). Individuals with psychiatric disorders such as obsessive-compulsive disorder (OCD), eating disorders, and anxiety disorders often display these maladaptive traits of perfectionism (Fairburn et al., 1999; Fairburn et al., 1997; Hewitt & Flett, 1991; Steele, Corsini, & Wade, 2007). These traits could be due in part to a maladaptive response system, or an inability to respond to errors and adjust subsequent behavior (Clayson & Larson, 2011; Schrijvers, De Bruijn, Destoop, Hulstijn, & Sabbe, 2010).

Using electroencephalogram (EEG) technology, the response system can be measured in the brain during two event related potentials (ERPs): the error-related negativity (ERN) and the feedback-related negativity (FRN). The ERN occurs 50 to 80 milliseconds after an error has occurred (Gehring, Goss, Coles, Meyer, & Donchin, 1993) while the FRN occurs 300 milliseconds after the receipt of negative feedback (Horan, Foti, Hajcak, Wynn, & Green, 2012). Although current research has shown perfectionism to be associated with higher (i.e., more negative) ERN amplitudes (Stahl et al., 2015), no studies have yet investigated the perfectionist’s neural response to negative feedback. The FRN amplitude, however, is blunted in OCD, schizophrenia, depression, and anxiety though—all of which have components of perfectionism.

We hypothesized that individuals with perfectionistic tendencies would exhibit larger ERN amplitudes and smaller FRN amplitudes than those who do not have such tendencies. To our knowledge, our study is the first to investigate how the FRN is affected by perfectionism and will help solidify the research concerning ERN and perfectionism due to our large sample size. Discovering the relationship of perfectionism and the response system could help in understanding the role perfectionism plays in more severe psychopathologies such as OCD and anxiety disorders.

Methodology

Participants wore a 128-electrode sensor EEG net to measure ERPs during two separate tasks: a flanker and a gambling task. This first task consisted of a modified Eriksen Flanker task to elicit the ERN (Eriksen & Eriksen, 1974). In this task, the participants were show five arrows and had to indicate the direction of the middle arrow. This middle arrow was either congruent (<<<<<) or incongruent (<<><<) with the surrounding arrows. The amplitude of the ERN was influenced by correct or incorrect identification of the middle arrow during each trial. This data, as well as reaction time (RT) and accuracy, were recorded on a separate computer.

The second task measured the FRN of participants through a simple gambling task (Horan et. al, 2012). Participants were shown a graphic of two doors and told to select which one they wanted to open. After choosing, participants were shown a green up arrow if they had guessed correctly and gained money ($.80) or a red down arrow if they had guessed incorrectly and consequently lost money ($.40). Participants were told that they would gain $.80 each time they opened a door that hid a prize or lose $.40 if they chose incorrectly, and that they would earn a sum between $0 to $20 at the end of the task. Feedback order was randomized across 50 trials for each participant with exactly 25 wins and 25 losses, resulting in a net sum of $10.00.

A total of 174 participants (91 female) performed the flanker and gambling task, with an age range of 18-25. Participants were predominately right handed (98.3%) and were free of any neurological diseases, psychiatric illnesses, learning disabilities, or head injuries that resulted in unconsciousness. After eliminating scores that were worse than chance, 104 participants (58 female) were used for the ERN sample and 166 participants (67 female, one unreported) were used for the FRN sample. All participants completed the Frost Multidimensional Perfectionism Scale and also completed the Beck Depression Inventory, Second Edition, and the State-Trait Anxiety Inventory to screen out potential confounds. For our statistical analyses, we used multiple regression with sex, age, depression/anxiety scores and perfectionism scores as independent variables and ERN and FRN amplitude as the dependent variables.

Results

As expected for the flanker task, incongruent trials were significantly associated with increased RT and decreased accuracy compared to congruent trials (ps < .01). Similarly, ERN and FRN amplitudes were more negative on error trials than correct trials (p<.01), indicating that the flanker task was successful in eliciting error responses and that the gambling task was successful in creating negative feedback. Correlation and regression analyses showed no significant relationships between ERN or FRN amplitude and perfectionism scores.

Discussion

Contrary to our hypothesis, participants with perfectionistic tendencies were not associated with increased ERN amplitude or decreased FRN amplitude, indicating that these perfectionistic tendencies did not adversely affect the response system. Our results imply that for psychological disorders, particularly OCD and anxiety disorders, perfectionism may not be one of the main factors in modifying neural indices of error monitoring.

Conclusion

Our hypothesis was that those with more perfectionistic tendencies would exhibit increased ERN and decreased FRN amplitudes. Our results did not support our hypotheses, showing no significant relationship between perfectionistic tendencies and neural indices of internal error monitoring. However, future studies may wish to examine more severe tendencies of perfectionism in those with psychological disorders to see if the severity of perfectionism is related to modified neural indices of error monitoring.