The relationship between anxiety and error-related negativity across development
Clinical anxiety that is characterized by persistent “worrisome thoughts, physiologic arousal, and strategic avoidance behaviors” (Moser,2013, 2) is one of the most widespread mental illnesses in the United States. Because of this rampant mental health problem, conducting research that examines the associations between anxiety and cognitive functioning is important in finding solutions to problems stemming from pathological anxiety (Moser, 2013, 2).
One area of research in the neuroscience field studies this issue by examining how anxiety is related to error monitoring, which is described as “the signaling and detection of errors in order to optimize behavior across a range of tasks and situations” (Moser, 2013, 3). The most common method used to measure the event-related potential (ERP) – the response that is the result of a specific event – is through an electroencephalogram (EEG) which noninvasively records the brain’s electrical activity at the scalp (Hajcak, 2016, 2). Error-related negativity (ERN), a “sharp negative-going ERP” (Hajcak, 2016, 2), occurs when an error is made by an individual. The ERN is most present 50 milliseconds after the error occurs. The ERN is thought to be generated in a region of the prefrontal cortex, specifically the anterior cingulate cortex, which is sensitive to “errors, pain and punishment” and modifies behavior (Hajcak, 2016, 3).
A common way of finding the ERN during an EEG is through a task on the computer referred to as “flankers”, a set of trial where the participant is to click on the right or left mouse button in accordance with right or left facing arrowheads on the screen (Hajcak, 2016, 1). When the participant presses the incorrect button, the ERN is present, even when the individual is not aware they have made a mistake. The size of the ERN is a direct reflection of the cost, or value, of the error. As G. Hajcak puts it, “when errors are more costly – and when performance is being evaluated – the ERN is larger” (Hajcak, 2016, 3).
The highlighted rectangle in Figure 1 shows the event-related potential when there is a correct and incorrect response to the computer task. The solid black line is the error-related negativity at its maximal, 50 milliseconds after the error occurred.
Studies that look at this relationship between the error-related negativity and anxiety have found that clinical anxiety is associated with an increased amplitude of the ERN, in both adults and older children. However, studies done on adolescents have found that anxiety is associated with a smaller ERN in younger children. Somewhere between the ages of six-to-ten years old, the amplitude of the ERN flips in children with diagnosed anxiety. The relationship between anxiety and the error-related negativity changes throughout development. This research is vital as it could possibly increase understanding, intervention, and prevention (Hajcak, 2016, 13).
A study by William J. Gehring and his colleagues was performed on 26 youth with Obsessive-Compulsive Disorder (OCD), 13 youth with non-OCD anxiety and 27 non-anxious youth from 8-to-16 years old (M. Carrasco et al., 2013, 1). The study measured the ERN during an Eriksen flankers test, recording that patients with OCD or non-OCD anxiety had a higher ERN amplitude. This study showed that an enlarged ERN is not limited to OCD (M. Carrasco et al., 2013, 2).
Though the amplitude of the ERN was significantly larger in patients with an anxiety disorder compared to the control group (non-anxiety patients), there was no significant difference in the amplitude in patients who were taking a serotonergic antidepressant or receiving cognitive-behavior therapy (M. Carrasco et al., 2013, 3). The study even found that children with successfully treated OCD with symptoms under the scores necessary to be diagnosed had an increased ERN. This finding infers that a larger ERN is a biomarker for OCD and other anxiety disorders (M. Carrasco et al., 2013, 2). Future studies could extend upon these findings and identify if other forms of therapy and treatments were effective in lowering the ERN in patients with a diagnosed anxiety disorder.
Figure 2 represents the three groups Gehring performed the study on- OCD youth, non-OCD youth, and the control group. As the first two graphs show, the anxiety disorder youth had a much larger amplitude.
Sharon L. Lo performed a study on 139 children (five-to-eight years of age) of which the purpose was to “examine the association between the ERN and diagnostically-defined symptoms of different anxiety disorders” (Lo, S. L., 2017, p. 1440), or the relationship between specific symptoms of different anxiety disorders and the ERN.
An interesting result recorded from this study showed that children with a higher level of Separation Anxiety Disorder (SAD) symptoms were associated with a lower ERN. And that children with lower levels of SAD symptoms recorded a higher amplitude of the ERN. Because of those results, the experimenters concluded that children with a high level of SAD symptoms “do not exhibit an ERN-behavior pattern that is indicative of effective error-monitoring functioning” (Lo, S. L., 2017, p. 1444).
Research has found that clinical anxiety follows a “developmental pathway beginning early in life”, meaning negative results in the testing of behavior in toddlers follows a path into adolescence and into adulthood, such as infants who “react negatively to novel stimuli” (A. Meyer, 2015, 2). Because of these findings, it is useful to perform longitudinal studies that examine the “neurodevelopmental correlates of specific cognitive processes that have been linked to anxiety” (A. Meyer, 2015, 1), or certain cognitive functions and processes that can be recorded that have been found to be associated with clinical anxiety. Doing so could further awareness of the possibility of anxiety disorders in older adolescents.
While no one study has examined the reasoning behind the almost opposite ERN amplitudes in young and older children, there is some speculation as to why it does so. As prior research demonstrates, the ERN most likely is generated in the anterior cingulate cortex (ACC) (A. Meyer, 2015). However, research also suggests that the ACC matures over the course of development (A. Meyer, 2015). As the ACC changes with age, as does the ERN. These studies have found a “significant correlation” between the volume and age of the ACC during computer tasks (A. Meyer, 2015). More specific studies examining the association between the ERN and anxiety as a “function of age” – which in turn would examine the possible reasons for the change in the ERN – could be very useful (A. Meyer, 2015).
The error-related negativity is considered a “neural biomarker” of anxiety, which is why neuroscience research focuses on it heavily. As previously stated, anxiety disorders are prevalent throughout the entire United States, so being able to predict the onset of them – by identifying biomarkers such as the ERN – is of great interest (A. Meyer, 2017). Identifying these biomarkers could improve the ability to prevent symptoms before they become debilitating, meaning new treatments in cognitive and pharmacological approaches becoming available (A. Meyer, 2017, 4).
Research suggests that as adolescents develop, their fears do not necessarily increase or decrease, rather they change from fear of external threat to internal threat (A. Meyer, 2017, 7). This could explain the shifting in the ERN amplitude throughout adolescence. However, more research needs to be performed to make this absolute conclusion.
A study performed by A. Meyer on 236 “healthy” children (no anxiety disorder at time of study) recorded the ERN at age six and then three years later at age nine. The purpose of the study was to figure out if an enhanced ERN could possibly predict the onset of an anxiety disorder later in life (A. Meyer, 2015, 1). The study found that an enhanced ERN at the age of six “predicted the onset” of an anxiety disorder at the age of nine. Not only that, there was a significant association between “maternal lifetime anxiety disorder” and the onset of a child’s anxiety disorder. This was the first study performed that has recorded that “enhanced error-related brain activity precedes the onset of anxiety disorders” (A. Meyer, 2015, 34).
Anxiety disorders are the most common form of mental illnesses; the impairing symptoms of such disorders are felt by a multitude of people. Therefore, there is much interest in performing research that identifies characteristics of anxiety disorders. Doing so could pave the way for treatments in the early stages of anxiety disorders.
Because anxiety begins early in development, current research has started to study developmental pathways that lead to anxiety in children. Much work has focused on error-related negativity as a neural biomarker of anxiety (Meyer, 2017, 3). The most recent studies have involved performing electroencephalograms on children with clinical anxiety over a certain period of time and recording the ERN as they complete tasks on the computer. The purpose of these studies has been to simply examine the relationship between the ERN and anxiety, and to observe its changes throughout development. To further that research, studies have been done to examine if the ERN could predict the onset of an anxiety disorder later in adolescence.
Many studies have recorded that the ERN has an increased amplitude in anxious adults. Although the same has been found in older adolescents, the opposite has been found in young children. Somewhere during development, this changes, yet no study has found reasoning for why that occurs. Some researchers have speculated that the reason for the near flip in amplitude is due to the maturation of the anterior cingulate cortex which is where the ERN is generated.
Since the ERN can predict the onset of anxiety disorders (A. Meyer, 2015, 34), continuing research can study new cognitive and behavioral treatments for these disorders. This is vital research that could eventually lead to the eradication of anxiety disorders.