Psychogenic Movement Disorders (PMD)

Psychogenic Movement Disorders (PMD) present a dilemma to neurologists and psychologists alike. Officially categorized as a Medically Unexplained Symptom (MUS), PMD manifests itself most commonly as a conversion disorder. A conversion disorder with psychogenic movements is defined by the DSM-IV as a movement disorder that cannot be explained by organic damage to the nervous system; conversion disorders are thought to have a psychological origin, and in contrast to factitious disorder or malingering, the symptoms are not intentionally produced or feigned. Because of the broad nature of this definition, PMD serves as an umbrella term for a variety of movement disorders, the most common being tremor, myoclonus, dystonia, parkinsonism, and seizure. Despite previous neglect by many clinicians, the disorder is receiving greater attention, and PMD patients are estimated to account for 2-3% of all patients in movement disorder clinics.

While there are several competing theories, PMD is hypothesized to have a basis in faulty inhibitory circuits of motor control. Additionally, the intensity of the psychogenic movements worsen when patients are exposed to stressful and/or emotionally-charged situations. A recent study investigated the role of emotional affect on the startle eyeblink reflex in the PMD population. Researchers found that unlike the healthy controls who displayed significant potentiation only for negative affect, psychogenic patients displayed significant potentiation of the startle response for both negative and positive pictures. While this study was still in only the preliminary stages due to its small sample group, the results indicate an abnormal modulation of physiological responses by emotional affect.

We chose to elaborate on this study by examining the role of emotional affect on motor inhibition, using the Stop-Signal Task (SST) as a measure of response inhibition. The Stop-Signal Task is commonly used to evaluate a subject's ability to suppress a pre-potent response, and the subtleties of the task are well-documented in the background literature. The paradigm allows researchers to discretely examine and compare the speed of both the motor and inhibition processes through the 'race-horse' model, which models motor initiation and inhibition as two competing mechanisms. Previous studies have shown the abnormal and/or impaired SST performance in individuals with Attention Deficit/Hyperactive Disorder and schizophrenia.

The SST itself consists of two types of tasks: a go task and a stop task. The two types are randomly intermixed within a single session, with a larger ratio of "go" trials to "stop" trials. In the "go" task, subjects fixate upon a cross which turns into a circle after a randomized time-interval. The circle remains on screen for a period of 1 second, and serves as an imperative "go" stimulus. Subjects are instructed to quickly press a key in response to the cue; all key-presses that occur outside the 1 second interval of the cue-appearance are logged as errors.

The "stop" task is almost identical to the "go" task. However, in the stop trials, an additional red "X" appeared after the circular go cue. The subjects were instructed to withhold all key presses at appearance of this "X". A successful stop-trial (SS) was recorded when subjects made no key-press while a stop-trial was recorded as a failure (SF) if the subjects made a key-press even after the appearance of the stop-cue. In the stop task, the length of time between the go-cue and stop-cue appearances is referred to as the Stop-Signal Delay (SSD). Clearly, the stop-task becomes easier as the SSD decreases and the subjects are quickly alerted to withhold their responses; similarly, at longer SSDs, the task becomes more difficult as the stop-trials become almost indistinguishable from the go-trials, leading to a higher percentage of stop-task errors.

Many studies use a "stair-case procedure" to dynamically adjust the SSDs throughout the stop-signal paradigm; by increasing the SSD by 50 ms after a successful stop trial and similarly decreasing the SSD by 50 ms after a failed stop trial, researchers are able to reach a "critical" time period when subjects are able to successfully withhold their responses in 50% of all stop trials. By calculating this critical SSD, we can then compute a value termed the Stop-Signal Reaction Time (SSRT) by subtracting the SSD from the mean go-trial Reaction Times. As defined by the previously described 'race-horse' model, the SSRT is a theoretical value that represents the amount of time needed for the inhibitory processes to successfully suppress a pre-potent motor response. The 50% success rate is critical for this calculation as it represents a state in which the competing motor initiation and inhibition processes are evenly matched. Thus, by comparing SSRTs across different patient groups and different manipulation, one can quantify the effectiveness of a subject's inhibitory processes, with shorter SSRTs representing better and more effective inhibition.

In our own study, we modified this task by adding an affective component. Prior to each trial of the stop-signal task, we displayed an affective image in order to examine the interactions between emotional affect and motor inhibition. The images themselves were chosen from the International Affective Pictures database, with forty-five images chosen in all. The affects were grouped by affective valence as either neutral (N), positive (P), or negative (U), with fifteen affective images per condition. In our modified task, subjects indicated their attention and recognition of the picture by responding whether or not the picture took place indoors or outdoors.

It is crucial to distinguish whether or not the errors made on the stop-trials are due to the natural progression of the stop-signal paradigm or a result of the affect; therefore, we essentially created three separate SSTs - one for each affective condition - and intermixed them throughout one session. The three SSTs operated completely independently of one another, with a stop-error for one condition having no effect on the dynamically-determined SSDs of the other two conditions.

Our modified stop-signal paradigm had a ratio of two go trials per every stop trial, with one trial per affective image for a total of 135 trials per a single session. Because the nature of the SST requires a minimum of 30 stop-trials to reach the critical 50% stop-success rate, each subject repeated four sessions for a total number of 540 trials, 180 of which were stop-trials. Because we wished to examine the neural correlates of the SST, we ran this paradigm simultaneously during a functional Magnetic Resonance Imaging scan.

Subjects
We tested this procedure on six patients diagnosed with Psychogenic Movement Disorder and seven age- and gender-matched healthy volunteers. All participants were screened for MRI safety compatibility. Because antidepressant medications have been reported to interfere with affective response, potential participants were excluded if they reported current antidepressants use. At the end of the experiment, participants were provided with monetary compensation.

Results
Because our sample size for both healthy and PMD populations is relatively small, we were forced to use less exacting statistical measures for significance, with less-stringent p-values and F-scores. As such, these reported findings represent only preliminary results. As we recruit more subjects, our data and results will become definitive.

Behavior
We focused much of our analysis on the SSRTs of both tested populations, as the SSRTs are well-documented as accurate estimators of motor inhibition ability. In the control population of the healthy volunteers, the average SSRT for the neutral condition was 216.20 ms, while the positive and negative conditions reported lower SSRTs of 198.69 ms and 204.76 ms respectively. The difference between the positive and negative conditions was determined to be not statistically significant by a t-test (p=0.77).

The PMD population showed a similar trend, with the positive affect having the lowest mean SSRT of 184.57 ms, the negative affect with a mean SSRT of 205.79 ms, and the neutral affect with the highest mean SSRT of 208.58 ms. While a t-test between the positive and negative conditions revealed that the difference between the two results was not statistically significant (p=0.155), it was indicative of a trend. We predict that with a greater number of subjects, the results will test as statistically significant.

As mentioned previously, a shorter SSRT is indicative of more efficient motor-inhibition processes. Thus, the behavioral trend, as consistent across both the PMD and healthy volunteer population, indicates that the inhibitory processes are actually facilitated by both positive and negative emotional affect. Similarly, a comparison between the two populations reveals that the PMD patients actually seem to have faster inhibitory processes than the controls for the positive and neutral affects, with no real different for the negative affects. These results are surprising, but not contradicted by the literature and worthy of greater investigation.

Imaging
We extracted the neural correlates involved in successful stop-task inhibition by contrasting the data for successful stop trials versus failed stop trials, for each individual emotional condition. By focusing on the results for the neutral affect, we were able to isolate the correlates of pure motor inhibition without interference from the emotional affect. Interestingly, we found that the healthy volunteers showed greater activation in the right inferotemproal cortex, the bilateral putamen, and the right amygdala. Previous studies have verified the importance of the right inferotemporal cortex in successful movement inhibition, and the putamen has a well-established role in motor control.

In addition, contrasts designed to yield brain activation correlated to negative or fearful emotional affect revealed that the healthy volunteer population showed greater activation in the right amygdala than the PMD patients. Studies have shown that the amygdala plays a vital role in processing of emotion.

As stated previously, these results are only preliminary and there are still several other components to compare in both the behavioral and functional imaging arms of our study. However, the current results are extremely promising and underscore the potential differences between the psychogenic movement and healthy volunteers populations. We hope that the further investigation will help reveal the workings of this mysterious disorder.