Exploring the correlation between limbic, cognitive and motor behavior in
healthy subjects
Student: Sanne Zomer Student number: 12140406 Supervisor: Bernadette van Wijk Submission date: 20th December 2020
Abstract
The subthalamic nucleus (STN) is an important structure when it comes to limbic, cognitive and motor behavior. This research is part of a study where the overall goal is to identify the amount of overlap between the cortico-basal ganglia networks involved in limbic, cognitive and motor behavior. This study can contribute to a better understanding of the functional organization of the STN and the workings of the cortico-basal ganglia networks. This can help with the further optimization of the DBS treatment in Parkinson patients and other clinical procedures. In this experiment, seven different tests, that are designed to measure the aforementioned aspects of behavior, were executed. These tests are: an emotion recognition task, an approach avoidance task, a finger tapping task performed with both dominant and non-dominant hand, a verbal fluency task, solving math problems, and a working memory task. The reaction times and accuracy scores were analyzed for each task. We did not find evidence for a strong correlation between limbic, cognitive and motor behavior. Strong correlations were found between the motor tasks with the dominant and non-dominant hand, and between the emotion recognition task and the approach avoidance task. These correlations are from tasks that measure the same modality, motor behavior and limbic behavior, respectively. However, the correlations that were found between modalities indicate that there is a link between the limbic, cognitive and motor behavior in healthy subjects.
Key words: Subthalamic nucleus, basal ganglia, Deep brain stimulation, Parkinson’s disease Introduction
Parkinson’s disease (PD) is a progressive neurodegenerative disorder that is associated with cell death of dopaminergic neurons in the substantia nigra, and the aggregation of abnormally folded α-synuclein proteins, (Kalia & Lang, 2015). The dopamine deficit that occurs as a result can eventually leads to severe motor symptoms (Rizzi & Tan, 2017). These cardinal PD symptoms are bradykinesia, tremor, rigidity, and gait disturbance (Wong et al., 2019). The symptoms do not have to be present from the onset of the disease. Hoehn & Yahr (1967) categorized PD in five different stages. These stages range from the least to worst amount of symptoms. Symptoms like tremors; repeated oscillating movements, rigidity and difficulty moving on one side of the body, occur at the onset of the disease. In the fifth stage, the patient is bound to its bed or wheelchair due to balance disorders and the severe tremors on both sides of the body. The scale of Hoehn & Yahr (1967) includes only the motor symptoms of PD. It does not take into account non-motor problems associated with PD, such as impaired executive function, depression or apathy (Goetz et al., 2004). The frequency of these individual non-motor symptoms in Parkinson’s disease vary from <10% to over 50% (Goldman & Postuma, 2014).
Dopamine-based therapies, such as levodopa, form the standard medical treatment for patients in the early stages of PD (Kleiner-Fisman et al., 2006). Verschuur et al. (2019) however, found that in patients with early PD, levodopa-based treatments had no disease-modifying effect. As the disease progresses, the dopaminergic-based therapies could become insufficiently effective or the side effects like dizziness and nausea, could become too severe. In this case, when a patient has motor fluctuations, periods when the medicine wears off and the symptoms come back, and levodopa induced dyskinesias as a side effect, the patient may be eligible for deep brain stimulation (DBS) treatment (Benabid, 2003).
DBS is a technique that consists of electrical stimulation via electrodes implanted deep inside the brain. The main target for implanting the DBS is the subthalamic nucleus (STN). The STN is part of the basal ganglia system. It is located ventral to the thalamus and dorsal to the substantia nigra and plays a role in the functional control of motor activity in the basal ganglia (Benabid, 2003).
There is a popular assumption that the primate subthalamic nucleus has three functional sub-regions. The limbic, associative and sensorimotor regions, laying in the anterior, mid and posterior STN, respectively (Joel & Weiner, 1996). However, the anatomical internal structure of the STN is still under debate. The variability across studies with regard to the amount of subdivisions and their localization is large (Keuken et al., 2012).
Although DBS can be very effective in suppressing motor symptoms, there are also associated risks. Neuropsychiatric side-effects like hypomania, acute sadness, impulsive or aggressive behavior, hilarity, or mania after DBS treatment are consistently reported (Benabid, Chabardes, Mitrofanis, & Pollak, 2009). With the placement of the electrodes, it is unavoidable that a subset of the electrode contacts is not perfectly placed in the putative motor part of the STN. The main reason for this is that, at present, the typical contacts leads are still larger than the motor part of the STN (resp ± 8 and 3 mm). This could cause the possible side-effects. Besides that, it is difficult for neurosurgeons to implant electrodes with sub-millimeter precision with respect to the target. As soon as the skull is opened, the pressure on the brain changes and the structures shift a bit in location compared to the pre-operative MRI scan. Furthermore, it is conceivable that motor and non-motor networks are anatomically and functionally overlapping. In that case, stimulating the motor network might be very difficult without affecting non-motor functions. Van Wijk et al. (2020) argue that regional
specialization without sharply defined borders is likely most representative of the STN's functional organization.
This research is part of a study where the overall goal is to identify the amount of overlap between the cortico-basal ganglia networks involved in limbic, cognitive and motor behavior. This is important for the determination of the optimal placement of the DBS electrodes. The study will focus on the correlation between limbic, cognitive, and motor behavior in healthy subjects, to see if a possible overlap between anatomical networks is reflected in the performance scores on a number of simple experimental tasks. This work will act as a pilot study before continuing with Parkinson’s disease patients. Together, this research can contribute to a better understanding of the functional organization of the STN and the workings of the cortico-basal ganglia networks. This can help with the further optimization of the DBS treatment in Parkinson patients and other clinical procedures. The research question of this study is: to what extent are limbic, cognitive and motor behavior correlated in healthy subjects? The hypothesis is that if there is a large overlap between the anatomical networks of limbic, cognitive, and motor functions, this would be reflected in a correlation between the behavioral scores on experimental tests of the different modalities. For the overall research, this could mean that if there is a correlation, the STN might not consist of clearly segregable sub-regions. The side-effects due to the electrode placing are then probably more difficult to reduce than when there is no correlation between the different modalities, which would suggest that independent networks underlie limbic, cognitive, and motor functions.
Subjects in my study performed seven different tests that are designed to measure the aforementioned behavior. A correlation analysis was applied to reveal the relations between behavioral scores of the different modalities.
Alkemade & Forstmann (2014) suggested to revise the tripartite subdivision hypothesis of the human STN to it having an organization without strict anatomical boundaries. This discussion leads to the prediction that there is an overlap between the anatomical networks of the limbic, cognitive and motor functions. As a result, the prediction is that a significant positive correlation will be found in this study between limbic, cognitive and motor behavior in healthy subjects. The correlation between modalities is predicted to be less strong than the correlation within modalities.
Method Subjects
31 subjects were tested (6 male, 25 female, aged 18-26 (sd=1.49)).The subjects were recruited via the Psychology subject pool of the University of Amsterdam and via personal recruitment. Subjects were excluded from the study if they have a history of neurological or psychiatric problems. Besides that, the subjects needed to have good vision, possibly corrected by glasses or contact lenses. If the subject has dyscalculia, they were excluded. Subjects who have movement or coordination problems were also excluded from the study. Because of the verbal fluency test, this study was only accessible to Dutch subjects.
Location
The tests were, if possible, performed in one of the experimental rooms in building REC-L. If not, an alternative location, like a quiet room at home, was used. There had to be as little distraction as possible so that the subject can fully focus on the tasks. The participants were tested individually. Procedure
The project was approved by a member of the Ethics Review Board. The experiment starts with the subjects getting an explanation of what is going to happen and what is expected. They have to read and sign a consent form before the tests can be executed. The subject has to answer a few questions about their age, gender, handedness, and mental state at this particular moment. This will include questions about their motivation and concentration. After that, the first task will be presented. The experiment in total will take about 45 minutes.
There are seven tasks that measure different disciplines. In this study the focus is on limbic, cognitive and motor functions. There are tasks that are more focused on one specific modality, and there are tasks that involve multiple modalities. The tasks are given in a random order, to prevent biases based on concentration loss or fatigue. They were programmed in PsyPad. All the tasks were executed on an iPad. Screenshots of the different tasks are given in figure 1.
1. Multidigit Mental Multiplication task (cognitive)
To test executive functions, subjects have to solve mathematic problems by mental arithmetic. The Multidigit Mental Multiplication Task will be used in this experiment. There will appear a mathematic problem on the screen. Following the study of Han, Yang, Lin, & Yen, 2016, the presented math problems are 2 digits multiplied by 1 digit (2 × 1). There is a 15 second time limit on the screen-time of the 2x1 problems. When the time is up, the subject sees a black screen and then has 6 extra seconds to give their answer. After this, the next mathematic problem will appear. The subject can give their answer when they think they know the correct answer, even when the problem is still on the screen. The answers will be given verbally, so an audio-recorder will be used to analyze speed and the given answer. An auditory pulse is given at the moment the math problem is displayed for
synchronization between the stimuli presentation and the audio recording. Only the first given answer will be checked. The time until this given answer is specified as the reaction time.
In total, 14 problems will be tested. This task will take about 5 minutes. Only when the subject gives the correct answer in the given time, there will be points awarded.
The outcome variables are: reaction time, total number of correct answers. 2. Emotional recognition task (limbic)
An emotional facial expressions recognition task is used to test limbic function. This test uses a range of photographs from the Warsaw Set of Emotional Facial Expression Pictures (WSEFEP). The faces of 30 individuals (16 female, 14 male) were used, each displaying a neutral expression, and the six basic emotions, which are happiness, sadness, disgust, fear, surprise and anger. In this test, only the emotions disgust, fear, surprise and anger will be used. The subject gets to see a picture, randomly picked from this series. Below the picture are four buttons with the possible emotions written. The subject has to choose which emotion is best represented by the picture by pressing the button with the corresponding emotion as fast as possible. The reaction time is defined by the time the picture appears on the screen until one of the buttons is pressed. The subject must assess 120 pictures. There is no time limit on the judgement of a single picture. This task will approximately take 5 minutes.
The outcome variables are: reaction time, total number of correct answers. 3. Bradykinesia Akinesia Incoordination (BRAIN) test (motor)
An adapted version of the point-to-point finger tapping task will be performed to test motor function: the Bradykinesia Akinesia Incoordination (BRAIN) test (Noyce et al., 2014). The subject is instructed to alternately tap two blocks that are depicted on the outer left and outer right side of the screen as fast as they can until they reach 75 taps (Noyce et al., 2014). Each hand will be tested separately. The subject will be told which hand is first to be tested. This will be randomized between the subjects. The task will be performed with every hand once.
To measure the motor function, the accuracy of key presses, will be saved. This is done by
programming a bigger block, in the same color as the background, underneath the smaller blocks. In this way, if the subject taps next to the block, it will still be registered. Besides that, the reaction time will be saved. This task will take about 2 minutes.
The outcome variables are: accuracy (as a number), average time per tapping movement. 4. Verbal Fluency test (cognitive and motor)
The verbal fluency test will be used to measure executive functioning and motor function. This task consists of 2 parts: the letter-based fluency task and the semantic fluency task. These tasks will be counterbalanced between the subjects.
In the letter-based fluency task, participants will have to name as many words beginning with a given letter in one minute (Benton & Hamsher, 1976). It is not allowed to generate names of people or places. The test that is mostly used these days uses the letters F, A and S (Spreen & Strauss, 1988). This is based on the English vocabulary size per letter, where the F, A and S are classified as easy letters to form words with. In the Dutch language, the letters that are commonly used for this kind of test are D, A, T, K, O, M (Schmand, Groenink, & den Dungen, 2008). In this task, the letters D, A, and M will be used.
The other part of the task will be the semantic fluency task. The subject has to name as many objects from a certain category, in this case animals and sports, in one minute, while the answers were audio-recorded to analyze speed, fluency and possible mistakes. An auditory pulse is given at the moment the task is displayed for synchronization between the stimuli presentation and the audio recording. The time until the last given answer is specified as the reaction time.
This task will take about 5 minutes.
The outcome variables are: total of correct answers from the letter-based and semantic fluency task combined, average time per given answer.
5. Stroop task (cognitive and motor)
Another test to measure executive functioning and motor function is the Stroop task. In this task, words describing a particular color appear on the screen. There are two trial types. In congruent trials, the color of the word and the color described by the word are the same. In the incongruent trials, the words will appear in a different color than the color the word indicates (Stroop, 1935). The subjects must indicate the color of the word, not the color that the word describes. There will be 28 congruent trials, and 28 incongruent trials. The subjects have to press one of the four buttons describing the possible answers, then the next trial will appear. This task will take about 4 minutes. The outcome variables are: reaction time, total number of correct answers.
6. Word list working memory task (cognitive and limbic)
A working memory task will be used to test limbic function and executive functioning. The task includes an emotional, motor and neutral word list, all consisting of 20 words. The subject gets to see 20 words of one of the categories on the screen for forty seconds and has to remember these. After this time, the subject gets forty seconds to recall the words they just learned. An auditory pulse is given at the start of the recall for synchronization between the stimuli presentation and the audio recording. Their verbal answers were be recorded and scored afterwards. The time until the last given answer is specified as the reaction time. The order of the different list categories were randomized. This task will take about 5 minutes.
The outcome variables are: reaction time, total number of correct answers. 7. Approach Avoidance (AAT) task (motor and limbic)
An approach avoidance task (AAT) will be performed to tests motor and limbic function. In this task, a picture was presented on the screen. At that point, they have to respond as quickly as possible to each stimulus by tapping on a button with ‘POSITIVE’, that is situated under the picture, closer to the participant, or on a button with ‘NEGATIVE’ on it that is above the picture, further away from the participant. The theory behind this task is that pleasant stimuli would produce automatic approach tendencies, whereas negative stimuli produce automatic avoidance tendencies (Chen & Bargh, 1999). Approach is associated with pulling an object closer, whereas avoidance is associated with pushing objects away from themselves (Chen & Bargh, 1999; Solarz, 1960). 60 pictures, selected from the Open Affective Standardized Image Set (OASIS; Kurdi, Lozano, & Banaji, 2016) will be used. 30 negative pictures, and 30 positive pictures. The response is considered correct if the participant’s choice is in accordance with the category of the picture (positive or negative). This task will take about 4 minutes.
The outcome variables are: reaction time, total number of correct answers. Data analysis
To start with, a pair-wise correlation between the different outcome variables will be done. Based on the numbers from the Shapiro Wilk test, the Spearman correlation was performed when data did not meet the assumption of normality. A Pearson correlation was performed when the data did meet the assumption of normality. If one of the variables was normally distributed and the other not, a
The reaction time of the different tasks will be correlated with each other, just as the accuracy of the different tasks. Given the relatively large number of variables, a principal component analysis will be done to see if some variables share variance.
Figure 1. Screenshots of the different tasks from the experiment.
Results
For every performed task, the reaction time and accuracy of the 31 participants were analyzed. The mean scores and reaction times are given in table 1. An overview of the score and reaction time per subject for the different tasks are shown in figure 2 & 3.
Mean reaction time (s) Mean Score
Math problems 9.84 (sd = 2.37) 10.39 (out of 14, sd = 2.25)
Emotion recognition 1.77 (sd = 0.33) 99.00 (out of 120, sd = 7.19) Motor dominant 0.314 (sd = 0.07) 62.45 (out of 75, sd = 17.30) Motor non-dominant 0.387 (sd = 0.07) 66.13 (out of 75, sd = 8.68)
Verbal Fluency 3.23 (sd = 0.71) 84.19 (sd = 21.10)
Stroop 1.23 (sd = 0.25) 50.13 (out of 56, sd = 1.23)
Working memory 2.20 (sd = 0.53) 28.74 (out of 40, sd = 6.15) Approach Avoidance 1.13 (sd = 0.17) 57.68 (out of 60, sd = 1.62) Table 1. Mean reaction time and scores on the different tasks.
The Shapiro-Wilk test was used to check for normality. The reaction time data of the approach avoidance task, the verbal fluency task, the math problems and the working memory task were not found to significantly differ from a normal distribution (p=.055 p=.733, p=.112, p=.303, respectively). The reaction times for the other tasks were found to significantly differ from a normal distribution (p=.0101 for the emotion recognition task, p<.001 for the motor task, p<.001 for the non-dominant motor task, and p=.031 for the Stroop task). For the accuracy scores, the data of the approach avoidance task, the verbal fluency task, the math problems and working memory task were not found to significantly differ from a normal distribution (p=.055, p=.733, p=.112, p=.303, respectively). The accuracy data for the other tasks were found to significantly differ from a normal distribution (p=.011 for the emotion recognition task, p<0.001 for the motor task, p<0.001 for the non-dominant motor task, p=0.031 for the Stroop task).
The reaction times of the emotion recognition task significantly correlated with the approach avoidance task (ρ=0.72, p<.001). Besides that, the reaction times of the motor task significantly correlated with the non-dominant motor task (ρ=0.84, p<.001). Another significant correlation is found between the emotion recognition and the Stroop task (ρ=0.55, p<.005), and between the approach avoidance and Stroop task (ρ=0.50, p<.005). The emotion recognition task significantly correlated with the verbal fluency task (ρ=0.43, p=.018), and the approach avoidance significantly correlated with the verbal fluency task (ρ=0.43, p=.016). The correlations and p-values of the reaction time data are shown in table 2 and 3.
A strong significant correlation was found between accuracy scores for the motor and non-dominant motor task (ρ=0.87, p<.001). Other significant correlations in accuracy are found between the emotion recognition and motor task (ρ=0.46, p=.0101), between the approach avoidance task and the math problems (ρ=0.48, p=.0058), and between the verbal fluency and working memory task (ρ=0.37, p=.042). The correlations and p-values of the accuracy data are shown in table 4 and 5. A plot of some of the correlations that were found are shown in figure 4.
A Principal Component Analysis (PCA) showed which components explained most variance. The scree plot, which plots the eigenvalues from largest to smallest, is shown in figure 5. The first four
components were the most important in explaining the variability in the dataset (explaining 25.4%, 17.5%, 13.6% and 11.6%, respectively). The contribution of the variables to these components are shown in figure 6.
Discussion
Summarizing, there were a few interesting correlations found. For the reaction time, a strong significant correlation was found between the emotion recognition task and the approach avoidance task (ρ=0.72, p<.001), and between the motor and non-dominant motor task (ρ=0.84, p<.001). For the accuracy, a strong significant correlation was found between the motor and non-dominant motor task (ρ=0.87). These strong correlations are between tasks that measure the same modalities, in this case limbic and motor behavior. There were also some correlations cross-modalities found. For the reaction time, significant correlations were found between the emotion recognition and the Stroop task (ρ=0.55, p<.005), and between the approach avoidance and Stroop task (ρ=0.50, p<.005). The emotion recognition task significantly correlated with the verbal fluency task (ρ=0.43, p=.018), and the approach avoidance significantly correlated with the verbal fluency task (ρ=0.43, p=.016). For the accuracy, this were correlations between the emotion recognition and motor task (ρ=0.46, p=.0101), between the approach avoidance task and the math problems (ρ=0.48, p=.0058), and between the verbal fluency and working memory task (ρ=0.37, p=.042).
These results show correlations between tasks more concentrated on the limbic versus cognitive behavior, and between tasks more concentrated on limbic versus motor behavior.
The PCA analysis showed that two third of the variance in the dataset could be explained by four principal components. In PC1, the reaction time and accuracy of the motor task for both the
dominant and non-dominant hand contribute a lot to this component, just as the reaction time of the emotion recognition, the approach avoidance and the Stroop task. For PC2, the accuracy for the approach avoidance task and the motor task for both the dominant and non-dominant hand contribute a lot, even as the reaction time of the approach avoidance task. It is striking that the motor task and the reaction times of different tasks, which all have a motoric influence, contribute to the components explaining the most variability. This indicates that the reaction times, defining the motoric component of each task, overall explain the most variance in the data.
The research question of this study is: to what extent are limbic, cognitive and motor behavior correlated in healthy subjects? The strong correlations that are found in this dataset are between tasks that measure the same modalities. There was no indication for a strong correlation between limbic, cognitive and motor behavior. However, the correlations between modalities were expected to be less strong than within-modality correlations. The correlations that were found between modalities indicate that there is a link between the limbic, cognitive and motor behavior in healthy subjects.
As stated before, the STN is important for limbic, cognitive and motor behavior. The relationship between these modalities has already been investigated in previous research. There is a link found between limbic and cognitive behavior. It is purported that increased arousal can enhance cognitive processing by altering the signal-to-noise ratio of neurological systems. The enhanced processing can in turn enhance attention, stimulus selection, and decision making (McMorris & Hale, 2015). Strong emotions can be the cause of increased arousal, what gives a link between limbic and cognitive behavior. In this study, there were indeed correlations found between limbic and cognitive behavior. Besides that, there are a lot of studies demonstrating that emotion can influence perception and behavior. Both positive effects, like contrast sensitivity and search efficiency (Becker, 2009), and negative effects, e.g. impairing behavioral performance when the processing of the unpleasant stimuli is irrelevant for the task, could be a consequence of intense emotion during an experimental task (Pereira et al., 2010). This finding could also be confirmed by this current study, because of the correlations found between limbic and motor behavior.
The correlations that were found may differ slightly from the actual values. There are some reasons that would explain a different outcome. First of all, the circumstances during this study were not ideal because this study was done during the COVID-19 pandemic. Due to the pandemic, the number of student registrations for the experiment was lower than expected. Because of that, many
participants were tested outside the lab environment. Things like the environment and possible distractors while doing the tasks that are wanted to keep the same as much as possible in an
experiment, were now very variable. Overall, there was more distraction and less concentration with these participants.
Besides the variability in the testing environment, there was also an inaccuracy with the data analysis. We used audio pulses for the tasks that required verbal response. The tasks were
programmed in PsyPad, which uses seconds as time unit. It is uncertain how long it took for the iPad to generate te sound. This caused in unknown delay between the start of the trial and the
presentation of the auditory tone. For this reason, the reaction times from the dataset were not as accurate as it could be.
For further research, the experimental tasks can be done with elderly and Parkinson’s disease (PD) patients to see whether disease duration and deep brain stimulation (DBS) have an impact on this correlation. Previous research showed that age is a factor that can influence the interference of motor and cognitive behavior. Lindenberger et al. (2000) came with the finding that postural tasks like walking, balancing or running receive priority for processing resources due to the large costs of failure. Costs of failure are higher in older subjects, what makes their resource allocation even more biased. This is confirmed in a multitask condition where participants had to execute different tasks during simulated car driving. In the older aged group, multitasking was less successful than in the group with younger participants (Wechsler et al., 2018). The findings from this study would probably not hold for an older PD group. It would be interesting to see whether this is the case.
To sum up everything that has been stated, strong significant correlations that are found are between the motor and the non-dominant motor task, and between the emotion recognition task and the approach avoidance task. These correlations are from tasks that measure the same modalities, motor and limbic behavior, respectively. There are also various significant correlations between modalities found, which indicates a link between limbic, cognitive and motor behavior. Due to the unfortunate circumstances during the study, the results could deviate from the actual link between modalities.
Figure 2. Reaction time per subject for the emotion recognition task (mean = 1.73, sd = 0.33), the approach avoidance task (mean = 1.13, sd = 0.17), the motor task (mean = 0.31, sd = 0.07), the non-dominant motor task (mean = 0.39, sd = 0.07), the verbal fluency task (mean = 3.23, sd = 0.71), the math problems (mean = 9.84, sd = 2.37), and the working memory task (mean = 2.20, sd = 0.53). The data of the motor tasks, the math problems and the working memory task are normally distributed. The data of the emotion recognition task, the approach avoidance task, Stroop, and verbal fluency task are not normally distributed. On the x-axis, the different subjects are given. On the y-axis, the reaction time in seconds are given.
Figure 3. Score per subject for the emotion recognition task (mean = 99.00, sd = 7.19), the approach avoidance task (mean = 57.68, sd = 1.62), the motor task (mean = 62.45, sd = 17.30), the non-dominant motor task (mean = 66.13, sd = 8.68), the verbal fluency task (mean = 84.19, sd = 21.19), the math problems (mean = 10.39, sd = 2.25), and the working memory task (mean = 28.74, sd = 6.15). The data of the approach avoidance task, the verbal fluency, the math problems and the working memory task is normally distributed. The data of the emotion recognition task, the motor tasks, and the Stroop task are not normally distributed. On the x-axis, the different subjects are given. On the y-axis, the scores on the different tasks are given.
Table 2 Correlations between the reaction times of the different tasks. Numbers in bold are Pearson correlations, other numbers are Spearman correlations. The grey background shows the correlations that have a p-value less than the α-level of 0.05.
Table 3. P-values of the correlations between the reaction times of the different tasks. Numbers in bold are Pearson correlations, the numbers that are not in bold are Spearman correlations. The grey background shows the correlations that have a p-value less than the α-level of 0.05.
Math
Problems Emotion Recognition Motor Motor ND Verbal Fluency Stroop Working memory Emotion Recognition 0.033 - - - - - Motor dominant 0.656 0.271 - - - - Motor non-dominant 0.877 0.282 <0.001 - - - Verbal Fluency 0.954 0.018 0.591 0.349 - Stroop 0.368 0.001 0.208 0.147 0.590 - Working memory 0.551 0.116 0.635 0.736 0.605 0.475 Approach Avoidance 0.386 <0.001 0.181 0.337 0.016 0.004 0.114 Math
Problems Emotion Recognition Motor Motor ND Verbal Fluency Stroop Working Memory Approach Avoidance Math Problems 1.0 - - - - Emotion Recognition 0.03 1.0 - - - - Motor task -0.08 0.20 1.0 - - - - - Motor ND 0.03 0.20 0.84 1.0 - - - - Verbal Fluency 0.01 0.43 0.1 0.17 1.0 - - - Stroop 0.17 0.55 0.23 0.27 0.10 1.0 - - Working Memory 0.11 0.12 -0.09 -0.06 -0.13 -0.13 1.0 - Approach Avoidance 0.16 0.72 0.25 0.18 0.43 0.50 0.29 1.0
Table 4. Correlations between the accuracy of the different tasks. Numbers in bold are Pearson correlations, the numbers that are not in bold are Spearman correlations. The grey background shows the correlations that have a p-value less than the α-level of 0.05.
Math
Problems Emotion Recognition Motor Motor ND Verbal Fluency Stroop Working Memory Approach Avoidance Math Problems 1.0 - - - - - - - Emotion Recognition 0.29 1.0 - - - - - - Motor 0.31 0.46 1.0 - - - - - Motor ND 0.16 0.29 0.87 1.0 - - - - Verbal Fluency 0.13 -0.08 0.04 0.13 1.0 - - - Stroop 0.19 -0.25 -0.18 -0.15 0.04 1.0 - - Working Memory 0.06 -0.07 -0.20 -0.14 0.13 0.10 1.0 - Approach Avoidance 0.48 0.14 -0.01 -0.14 0.09 0.28 0.10 1.0
Table 5. Correlations between the accuracy of the different tasks. Numbers in bold are Pearson correlations, the numbers that are not in bold are Spearman correlations. The grey background shows the correlations that have a p-value less than the α-level of 0.05.
Math
Problems Emotion Recognition Motor Motor ND Verbal Fluency Stroop Working Memory Emotion Recognition 0.11 - - - - Motor 0.09 0.010 - - - - - Motor ND 0.38 0.11 <0.001 - - - - Verbal Fluency 0.49 0.65 0.84 0.47 - - - Stroop 0.32 0.18 0.34 0.42 0.83 - - Working Memory 0.74 0.69 0.29 0.46 0.042 0.58 - Approach Avoidance 0.006 0.47 0.97 0.46 0.62 0.13 0.60
Figure 4. Correlations between some of the found correlations. The significant correlation (ρ=0.84, p<.001) of the reaction time between the motor and non-dominant motor task is shown in this plot. On the x-axis, the reaction time of the motor task in seconds is given. On the y-axis, the reaction time of the non-dominant motor task in seconds is given. Furthermore, the significant correlations in accuracy that are found between the emotion recognition and motor task (ρ=0.46, p=.0101), between the approach avoidance task and the math problems (ρ=0.48, p=.0058), and between the verbal fluency and working memory task (ρ=0.37, p=.042) are shown in this figure. On the x- and y-axis are the scores for the different tasks given.
Figure 5. Scree plot of eigenvalues ordered from largest to smallest. The scree plot gives the percentages of explained variances per components. On the x-axis, the different components are given. On the y-axis, the percentages of explained variances are given.
Figure 6 Contribution of variables to PC1, PC2, PC3 and PC4 . Variables that are correlated with PC1, PC2, PC3 and PC4 are the most important in explaining the variability in the dataset. On the x-axis, the different variables are given. On the y-axis the percentage of contribution to the different components are given.
References
- Alkemade, A., Schnitzler, A., & Forstmann, B. U. (2014). Topographic organization of the human and non-human primate subthalamic nucleus. Brain Structure and Function, 220(6), 3075–3086. https://doi.org/10.1007/s00429-015-1047-2
- Barry, D., Bates, M. E., & Labouvie, E. (2008). FAS and CFL Forms of Verbal Fluency Differ in Difficulty: A Meta-analytic Study. Applied Neuropsychology, 15(2), 97–106.
https://doi.org/10.1080/09084280802083863
- Becker, M. W. (2009). Panic search: Fear produces efficient visual search
for nonthreatening objects. Psychological Science, 20, 435-437.
- Benabid, A. L. (2003). Deep brain stimulation for Parkinson’s disease. Current Opinion in Neurobiology, 13(6), 696–706. https://doi.org/10.1016/j.conb.2003.11.001
- Benabid, A. L., Chabardes, S., Mitrofanis, J., & Pollak, P. (2009). Deep brain stimulation of the subthalamic nucleus for the treatment of Parkinson’s disease. The Lancet Neurology, 8(1), 67–81. https://doi.org/10.1016/s1474-4422(08)70291-6
- Benton AL, Hamsher K. Multilingual Aphasia Examination manual. University of Iowa; Iowa City: 1976
- Chen M, Bargh JA. Consequences of automatic evaluation: Immediate behavioral
predispositions to approach or avoid the stimulus. Personality and Social Psychology Bulletin. 1999;25:215–224. https://doi.org/10.1177/0146167299025002007
- Delazer, M., Domahs, F., Bartha, L., Brenneis, C., Lochy, A., Trieb, T., & Benke, T. (2003). Learning complex arithmetic—an fMRI study. Cognitive Brain Research, 18(1), 76–88.
https://doi.org/10.1016/j.cogbrainres.2003.09.005
- Ekman, P., & Friesen, W. (1976).Pictures of facial affect. Palo Alto, CA: Consulting Psychologists Press.
- Goetz, C. G., Poewe, W., Rascol, O., Sampaio, C., Stebbins, G. T., Counsell, C., … Seidl, L. (2004). MovementDisorder Society Task Force report on the Hoehn and Yahr staging scale: Status and recommendations TheMovementDisorder Society Task Force on rating scales for Parkinson’s disease. Movement Disorders, 19(9), 1020–1028.
https://doi.org/10.1002/mds.20213
- Goldman, J. G., & Postuma, R. (2014). Premotor and nonmotor features of Parkinsonʼs disease. Current Opinion in Neurology, 27(4), 434–441.
https://doi.org/10.1097/wco.0000000000000112
- Grabner, R. H., Ansari, D., Reishofer, G., Stern, E., Ebner, F., & Neuper, C. (2007). Individual differences in mathematical competence predict parietal brain activation during mental calculation. NeuroImage, 38(2), 346–356. https://doi.org/10.1016/j.neuroimage.2007.07.041 - Han, C.-C., Yang, T.-H., Lin, C.-Y., & Yen, N.-S. (2016). Memory Updating and Mental
Arithmetic. Frontiers in Psychology, 7, 72. https://doi.org/10.3389/fpsyg.2016.00072 - Joel, D., & Weiner, I. (1996). The connections of the primate subthalamic nucleus: indirect
pathways and the open-interconnected scheme of basal ganglia-thalamocortical circuitry. Brain Research Reviews, 23(1–2), 62–78. https://doi.org/10.1016/s0165-0173(96)00018-5 - Kalia, L. V., & Lang, A. E. (2015). Parkinson’s disease. The Lancet, 386(9996), 896–912.
https://doi.org/10.1016/s0140-6736(14)61393-3
- Keuken, M. C., Uylings, H. B., Geyer, S., Schäfer, A., Turner, R., & Forstmann, B. U. (2012). Are
there three subdivisions in the primate subthalamic nucleus?. Frontiers in neuroanatomy, 6, 14. https://doi.org/10.3389/fnana.2012.00014
- Kleiner-Fisman, G., Herzog, J., Fisman, D. N., Tamma, F., Lyons, K. E., Pahwa, R., Lang, A.E. & Deuschl, G. (2006). Subthalamic nucleus deep brain stimulation: Summary and meta-analysis of outcomes. Movement Disorders, 21(S14), 290–304. https://doi.org/10.1002/mds.20962
- Kurdi, B., Lozano, S., & Banaji, M. R. (2016). Introducing the Open Affective Standardized Image Set (OASIS). Behavior Research Methods, 49(2), 457–470.
https://doi.org/10.3758/s13428-016-0715-3
- Lambert, C., Zrinzo, L., Nagy, Z., Lutti, A., Hariz, M., Foltynie, T., … Frackowiak, R. (2012). Confirmation of functional zones within the human subthalamic nucleus: Patterns of connectivity and sub-parcellation using diffusion weighted imaging. NeuroImage, 60(1), 83– 94. https://doi.org/10.1016/j.neuroimage.2011.11.082
- Lang, P. J., Bradley, M. M., and Cuthbert, B. N. (2008). International Affective Picture System (IAPS): Affective Ratings of Pictures and Instruction Manual. Technical Report A-8.
Gainesville, FL: University of Florida.
- Lezak, M. D., Howieson, D. B., Bigler, E. D., & Tranel, D. (2012). Neuropsychological Assessment (5th edition). Oxford, UK: OUP USA.
- McMorris T., Hale B. J. (2015). Is there an acute exercise-induced physiological/biochemical threshold which triggers increased speed of cognitive functioning? A meta-analytic
investigation. J. Sport Health Sci. 4 4–13. https://doi.org/10.1016/j.jshs.2014.08.003
- Myerson, J., & Green, L. (1995). Discounting of delayed rewards: Models of individual choice. Journal of the Experimental Analysis of Behavior, 64(3), 263–276.
https://doi.org/10.1901/jeab.1995.64-263
- Noyce, A. J., Nagy, A., Acharya, S., Hadavi, S., Bestwick, J. P., Fearnley, J., … Giovannoni, G. (2014). Bradykinesia-Akinesia Incoordination Test: Validating an Online Keyboard Test of Upper Limb Function. PLoS ONE, 9(4), e96260.
https://doi.org/10.1371/journal.pone.0096260
- Odekerken, V. J. J., van Laar, T., Staal, M. J., Mosch, A., Hoffmann, C. F. E., Nijssen, P. C. G., … de Bie, R. M. A. (2013). Subthalamic nucleus versus globus pallidus bilateral deep brain stimulation for advanced Parkinson’s disease (NSTAPS study): a randomised controlled trial. The Lancet Neurology, 12(1), 37–44. https://doi.org/10.1016/s1474-4422(12)70264-8 - Pereira, M.G., de Oliveira, L., Erthal, F.S. et al. Emotion affects action: Midcingulate cortex as
a pivotal node of interaction between negative emotion and motor signals. Cognitive, Affective, & Behavioral Neuroscience 10, 94–106 (2010).
https://doi.org/10.3758/CABN.10.1.94
- Plummer P., Eskes G. A., Wallace S., Giuffrida C., Fraas M., Campbell G., et al. (2013).
Cognitive-motor interference during functional mobility after stroke: state of the science and implications for future research. Arch. Phys. Med. Rehabil. 94 2565–2574.
https://doi.org/10.1016/j.apmr.2013.08.002
- Rizzi, G., & Tan, K. R. (2017). Dopamine and Acetylcholine, a Circuit Point of View in Parkinson’s Disease. Frontiers in Neural Circuits, 11, 110.
https://doi.org/10.3389/fncir.2017.00110
- Schmand, B., Groenink, S. C., & den Dungen, M. (2008). Letterfluency: psychometrische eigenschappen en Nederlandse normen. Tijdschrift voor Gerontologie en Geriatrie, 39(2), 64– 74. https://doi.org/10.1007/bf03078128
- Solarz A. Latency of instrumental responses as a function of compatibility with the meaning of eliciting verbal signs. Journal of Experimental Psychology. 1960;59:239–245.
https://doi.org/10.1037/h0047274
- Spreen O, Strauss E. A compendium of neuropsychological tests: Administration, norms and commentary. 2nd edition Oxford University Press; New York: 1998.
- Stroop J. R. (1935). Studies of interference in serial verbal reactions. J. Exp. Psychol. 18, 643– 662. https://doi.org/10.1037/h0054651
- Wechsler, K., Drescher, U., Janouch, C., Haeger, M., Voelcker-Rehage, C., & Bock, O. (2018). Multitasking During Simulated Car Driving: A Comparison of Young and Older Persons. Frontiers in Psychology, 9, 910. https://doi.org/10.3389/fpsyg.2018.00910
- Wong, J. K., Cauraugh, J. H., Ho, K. W. D., Broderick, M., Ramirez-Zamora, A., Almeida, L., … Okun, M. S. (2019). STN vs. GPi deep brain stimulation for tremor suppression in Parkinson disease: A systematic review and meta-analysis. Parkinsonism & Related Disorders, 58, 56– 62. https://doi.org/10.1016/j.parkreldis.2018.08.017
