Cortical brain activity for Stressful decision-making
A research proposal about the effect of stress on decision -makingLars Woudstra
11762551
Examinator: dr. H.J. Krugers
Daily supervisors: dr. A.B. Mulder & dr. J.C. van Hooff
Senior researchers: dr. A.B. Mulder & dr. J.C. van Hooff
Abstract
Stress is a reaction of the body to events that threaten its homeostasis. It is considered to be unpleasant and it therefore affects people’s lifestyles. Making choices is one of many cognitive functions which are affected by stress. Previous research has shown that advantageous or disadvantageous decisions are marked in the prefrontal cortex by / and / brainwave ratios, respectively. This research proposal focuses on the effects of stress on decision making in involved brain areas. To examine this, a pilot study was executed in collaboration with the NS stationslab. This pilot study showed a significant greater decrease in both / and / ratio in the prefrontal cortex for participants who made disadvantageous decisions in a low stress condition, compared to participants who made disadvantageous decisions in a high stress condition. Therefore, it can be concluded that stress causes an increase in / and / ratio when disadvantageous decisions are made. Significant differences in alpha and theta waves between resting-state and the moment of decision-making, show another effect of stress on decision-making. Both alpha and theta powers decrease less under stressful conditions in the prefrontal cortex of disadvantageous decision-makers. Further research to these findings is necessary to bring new insights of the effect of stress on decision-making.
Summary
The effect of stress on decision-making has been researched for decades. However, the exact mechanism of cortical brain activity in brain areas involved in stressful decision-making is yet to be found. This is a research proposal to investigate how several brain areas work together under stressful circumstances in moments of decision-making. A pilot study has shown a significant difference in brainwaves at the moment of decision-making between advantageous and disadvantageous decisions in stress and no-stress conditions. Further research is necessary to reveal the mechanism of these brainwaves in the human brain.
Keywords
1. Introduction
1.1 Stress
During life, people have to cope with stress. People experience stress when they go to work or when they face difficult decisions. In the classical concept, stress is a physical and psychological reaction of the body to events that threaten its homeostasis (De Kloet, Joëls, & Holsboer, 2005). These events are known as stressors and do not necessarily have to be physical. Psychological stressors are even among the most powerful stressors for human beings (Heuser & Lammers, 2003).
According to research of De Kloet et al. (2005), stressors can cause increased arousal, alertness, vigilance, focused attention and cognitive processing. However, chronic stress can lead to long-lasting activations of these reactions and are therefore considered to be unpleasant (Taelman, Vandeput, Spaepen & Van Huffel, 2009). In addition, chronic stress can disturb cognitive processes such as decision-making (Brand, Labudda, & Markowitsch, 2006). Besides causing cognitive changes, stressors also cause physiological responses of the body by activating the sympathetic nervous system, which results in a release of noradrenaline and adrenaline (De Kloet et al., 2005). This physiological reaction will be discussed next in order to understand the basic mechanisms of stress.
1.2 The stress response
The body’s first reaction to stress is the release of catecholamines: noradrenaline and adrenaline (McEwen & Sapolsky, 1995). These represent the first wave of the stress response and cause arousal and alertness by increasing blood pressure and heart rate and allow a ‘flight-or-fight’ response (Goldstein, 1995).
Measuring heart rate variability (HRV) is a common used method to see fluctuations in heart rate while people are stressed. HRV is known to increase during rest and decrease during stress (Taelman et al., 2009). HRV measures are calculated from data collected from an electrocardiography (ECG) signal by defining the time between two consecutive R peaks. This is a so-called RR-interval. The variance in time between all RR-intervals in a recording gives the HRV assessed by time domain analysis (van Ravenswaaij-Arts, Kollee, Hopman, Stoelinga & van Geijn, 1993). A decrease of HRV means an increase in heart rate and is therefore an indicator for stress (Taelman et al., 2009).
The second ‘wave’ of a body’s response to stress starts with secretion of glucocorticoids. When a human experiences physical or psychological stress, the paraventricular hypothalamus (PVN) produces high amounts of vasopressin (VP) and corticotropin-releasing hormone (CRH). This, as the name suggests, activates the secretion of corticotropin (ACTH), which enters the blood circulation. ACTH then activates secretion of cortisol (in humans) in the adrenal gland (De Bellis et al., 1999). This is called the Hypothalamic-Pituitary-Adrenal (HPA) system (Munck, Guyre, & Holbrook, 1984).
Cortisol can cause several actions all over the body as it circulates in the blood. To prevent the system from overreacting, cortisol inhibits its own release by a negative feedback mechanism by binding to corticosteroid-receptors. The HPA system with its negative feedback mechanism is called the HPA-axis, which regulates the stress reaction in the human body (Heuser & Lammers, 2003).
As glucocorticoids and catecholamines induce a reaction in the entire human body, so does it in the brain. Research has been done to expose the effects of these hormones on cognitive function. Research has shown that glucocorticoids and catecholamines are involved
with memory formation and memory recall (Mendl, 1999), but also influence performance, judgement and decision-making (Staal, 2004). The exact mechanism of the effect of stress on these cognitive functions, is yet to be investigated. Therefore, this research proposal will be focusing on one of these cognitive functions: decision making.
1.3 Decision making
Researchers have questioned the underlaying mechanism of decision making for decades (Starcke, 2012). Most research is focused on the underlying brain areas and how they are affected by stress, using functional brain imaging techniques. For example, Brand and colleagues found out that working memory and executive function are cognitive processes involved in decision-making under stressful conditions (Brand et al., 2006). That’s why the most prominent neural correlate in stressful decision-making is thought to be the prefrontal cortex. Lesions in this brain area show a lack of ability to make well-founded decisions under explicit risk conditions. People who have these lesions prefer risky and disadvantageous options during moments of decision-making (Starcke, 2012). For example, people with lesions in the prefrontal cortex would choose a less profitable option in a game than people with a full functional prefrontal cortex.
All brain regions involved in decision-making under stressful conditions are the dorsolateral prefrontal cortex, the ventral prefrontal cortex, the anterior cingulate gyrus and the anterior cingular cortex, the orbitofrontal cortex, and the parietal cortex. The parietal cortex may explicitly be involved in decision-making under risk, because it is considered to be involved in estimating gains and losses (Starcke, 2012).
Electroencephalography (EEG) is an important method for finding out which and how several brain regions are working together in several cognitive functions (Subhani, Xia, & Malik, 2011). Research of Schutter & Van Honk (2005) has shown that the power of delta (1-3 Hz), theta (4-7 Hz) and beta (1(1-3-(1-30 Hz) frequencies are predictive values for advantageous and disadvantageous decision making. People with both higher / and / ratios made more disadvantageous choices in the Iowa gabling task than people with both lower / and / ratios. These slow wave/fast wave ratios are commonly used as markers for attention, but also seem to expose the mechanism of decision-making (Schutter & van Honk, 2005).
As for stress, strong evidence has been found for a correlation between mental stress and suppression of alpha waves and an increase in theta waves (Subhani et al., 2011), but also an increased beta activity in the prefrontal cortex of subjects with high stress levels compared to subjects with low stress levels (Saeed et al., 2015). According to this research, stress causes a difference in theta, alpha and beta waves in the prefrontal cortex. Therefore, / and / ratios in the prefrontal cortex might also differ as well as a consequence of stress in moments of decision-making. Research to the effect of stress on these ratio’s has to my knowledge never been done before. In this research proposal, the following question is raised: “What is the effect of stress on cortical brain activity in brain areas involved in decision making?” It was hypothesized that people who make disadvantageous choices under stressful conditions have both higher / and / frequency ratios in cortical areas involved in decision-making, than people who make advantageous choices. These frequencies are expected to be higher for people in high stress experimental conditions.
To examine this hypothesis, a pilot study has been executed by the NS (Dutch railway services). Understanding passengers’ choices and the stress they have to cope with, allows the NS to improve the overall train station experiences and eventually reduce stress among passengers.
2. Methods
2.1 NS-Stationslab
With the aim to investigate stress among train passengers, data for this experiment was provided by the ‘NS-stationslab’ in 2017. The NS-stationslab is a cooperation between the NS, ProRail (railway network), ‘Weekend van de wetenschap’ (annual Dutch science event) and the University of Amsterdam. In the NS-stationslab, passengers who traveled past ‘Amsterdam Centraal’ train station could join scientific experiments in their waiting time. Depending on the time they had, participants were randomly assigned to one of the three experimental conditions. All the experimental conditions were developed for the NS to improve waiting rooms or overall experience of train stations. All experiments were executed by a student research team led by dr. T. Mulder from the University of Amsterdam.
2.2 The task
The task used for this pilot study, was a virtual task for studying decision-making and stress. Participants were asked to image their selves into a virtual sketched scenario and make a decision based upon that scenario at a certain point in the task. Additionally, participants also got to see a video about the outcome of their decision. Prior to the task and afterwards, participants were asked to fill in questionnaires and their heart rate and blood pressure was measured.
The task started with an introduction video and a given scenario whether the participants were asked to imagine themselves going to a friend (low stress) or to a job interview (high stress) by train. Next, the participants were told how much time they had before the train left: 2, 8 or 25 minutes. Then, the participants were asked to make a choice between running or walking to their train, or shop on their way. This choice was made by pressing a button: ‘R’ for running, ‘W’ for walking and ‘S’ for shopping. A video of someone making their way to the train was shown after the participants made their choice. Finally, participants were shown the outcome of their choice: they saw a video of someone catching the train or someone who missed the train. A visualization of the task is seen in figure 1.
For the analysis, participants were divided into four groups depending on the choice they made in the task. Participants made an advantageous choice by (1) running to their train when they had two minutes left; (2) run or walk to their train when they had 8 minutes left; (3) run, walk or shop when they had 25 minutes left. The other choices were qualified as disadvantageous. Participants who were given the scenario of going to a friend were placed in a low stress group and participants who were given the scenario of going to a job interview were placed in a high stress group. Thus, there are four subgroups in a 2x2 experimental design based upon stress level and choice characteristic.
Figure 1: Visual representation of the experimental task in the NS stationslab pilot study 2.3 Data recording and analysis
In this study, data was collected from 63 healthy participants (38 female, 25 male), age in the range 16-92, with a mean value of 41.8 years.
During the experiment, EEG data was recorded with one single electrode on the forehead to measure brain waves in the prefrontal cortex, as the prefrontal cortex was considered as the brain region with the most correlation to stress and decision making (Starcke, 2012). A ground electrode was put upon the forehead and a reference electrode upon the ear. No other electrodes were added to keep the experiment as accessible as possible for participants to participate.
First, the collected EEG data was bandpass filtered between 0.5 and 48 Hz to filter out the mains grid (50Hz) and uninteresting frequencies below 0.5 Hz. After filtering, four participants were excluded due to measurement failures and empty data files. Another 32 participants were excluded due to unreliable determination of the moment of the decision.
The remaining 31 participants were divided in four groups. Participant data of these groups is presented in table 1. Within these groups, a power analysis was done in the four second time period where the decision is made and at 20 seconds of the introduction video. The power analysis was executed for delta (1-3 Hz), theta (4-7 Hz), alpha (8-12 Hz) and beta (13-30 Hz) frequencies in the prefrontal cortex. Delta, beta and theta frequency powers were used to calculate a difference in / and / ratio and reveal the effect of stress on these ratios. Theta and alpha differences were used to determine stress levels at the moment of choosing and differences in stress level between the experimental groups.
Statistical analysis was executed by a one-way ANOVA test in R-studio. Differences are considered to be significant when post-hoc analysis between two groups give a p value smaller than 0.05. All data was normally distributed.
During the whole task, heart rate was measured with an ECG electrode on the wrist to determine HRV of each group. This was done by calculating the average time of all RR-intervals in a recording of a participant (mean RR) and calculating the deviation of every single RR-interval from the average RR-RR-interval length (mean RR). These deviations give the HRV at a certain time in the recording of each participant (van Ravenswaaij-Arts et al., 1993). To investigate if stress was present at the moment of choosing for all participants in an experimental group, the HRV values are plotted against the time for every group of participants. An average of these HRV values is calculated over every 10 points, giving a visual representation of the HRV of one experimental group.
Table 1: Participant data (number, age, gender) for the different groups Low stress - advantageous Low stress – disadvantageous High stress – advantageous High stress - disadvantageous Number of participants 13 3 10 5 Age range (Years) 20 – 92 37 - 55 18 - 69 20 - 56 Mean age (Years) 44.5 47.7 44.4 41 Gender (Male – Female) 2 M (15%) – 11 F (85%) 2 M (67%) – 1 F (33%) 4 M (40%) – 6 F (60%) 3 M (60%) – 2 F (40%)
3. Results
3.1 EEGThe / and / ratio differences between the base line measurement in the introduction video and the moment of choosing are presented in table 2. A one-way ANOVA was executed for both / (p = 0.000507, df = 3, F = 7.051) and / (p = 0.000127, df = 3, F = 8.478) ratio differences.
Table 2: / and / ratio differences between introduction video (baseline) and the moment of choosing for all experimental groups.
Low stress - Advantageous Low stress - Disadvantageous High stress - advantageous High stress - Disadvantageous / ratio difference -0,0206 ( = 0,0764) -0,0809 ( = 0,0308) -0,0444 ( = 0,0578) +0,0053 ( = 0,0463) / ratio difference -0,0264 ( = 0,0443) -0,0620 ( = 0,0229) -0,0220 ( = 0,0506) +0,0068 ( = 0,0451)
Negative values mean a decrease in ratio between the introduction video and moment of choosing. Positive values mean an increase in ratio.
Post-hoc comparisons showed that in the low-stress condition / decreased significantly more in the group that made a disadvantageous decision as compared to the group that made an advantageous decision (p = 0.022). / also decreased more in the high stress disadvantageous group as compared to the low stress disadvantageous group (p = 2.5E-04). A similar difference is seen in / ratio between the low stress disadvantageous group and the high stress disadvantageous group, where the low stress disadvantageous group decreases more (p = 4.3E-05). The post-hoc comparison also showed that the high stress advantageous group decreased more in / ratio than the friend disadvantageous group (p = 0.037).
Theta and alpha power differences between the introduction video and the moment of decision-making were used as marker for stress and are presented in Figure 1. Differences in theta and alpha power were investigated between 20 seconds of introduction video (baseline) and four seconds of the moment of decision-making. All groups decrease in theta power during the decision-making moment, compared to the resting state theta power in the
introduction video. A one-way ANOVA was executed to compare theta power between all experimental groups (p = 2.7E-06, df = 3, F = 12.89). Significant differences were found between four groups as shown in figure 2.
Figure 3 shows the difference in alpha power between the introduction video and the moment of decision-making, executed with a one-way ANOVA (p = 7.5E-06, df = 3, F = 11.64). Except for the high stress disadvantageous group, all groups have an increased alpha power. Significant differences were found between three groups as shown in figure 3.
3.2 Heart Rate Variability
The HRV values were plotted against the time for each group of interest. Figure 4, 5, 6 and 7 show these plots of HRV values. Every grey dot is one HRV value at a certain time of one participant. A floating mean is calculated over every 10 dots and is visualized in red. At t = 97 – 101, the moment of decision-making in the task is seen as a green bar at the x-axis. The HRV plots are used as a visual representation for indicating stress at a certain time in the task, especially for the time window of the decision-making moment.
Figure 2: Theta power increase in the prefrontal cortex between introduction video and moment of choosing. Significant differences are visualized. Differences in power ratio are given in V2/Hz.
Figure 3: Alpha power increase in the prefrontal cortex between introduction video and moment of choosing. Significant differences are visualized. Differences in power ratio are given in V2/Hz.
Figure 4: HRV during the task for the low stress, advantageous choice group. In red the average HRV calculated over every 10 points. Green is the time window of interest. An HRV value of 0 means no time difference between the RR-interval and mean RR at that specific moment.
Figure 5: HRV during the task for the low stress, disadvantageous choice group. In red the average HRV calculated over every 10 points. Green is the time window of interest. An HRV value of 0 means no time difference between the RR-interval and mean RR at that specific moment.
-0,15 -0,1 -0,05 0 0,05 0,1 0,15 0 20 40 60 80 100 120 140 160 180 200 H ea rt r at e v ar ia b ili ty ( s) Time (s) Low stress - Advantageous
-0,15 -0,1 -0,05 0 0,05 0,1 0,15 0 20 40 60 80 100 120 140 160 180 200 H e ar t ra te v ar ia b ili ty ( s) Time (s) Low stress - Disadvantageous
Figure 6: HRV during the task for the high stress, advantageous choice group. In red the average HRV calculated over every 10 points. Green is the time window of interest. An HRV value of 0 means no time difference between the RR-interval and mean RR at that specific moment.
Figure 7: HRV during the task for the high stress, disadvantageous choice group. In red the average HRV calculated over every 10 points. Green is the time window of interest. An HRV value of 0 means no time difference between the RR-interval and mean RR at that specific moment.
-0,15 -0,1 -0,05 0 0,05 0,1 0,15 0 20 40 60 80 100 120 140 160 180 200 H e ar t ra te v ar ia b ili ty ( s) Time (s) High stress - Advantageous
-0,2 -0,15 -0,1 -0,05 0 0,05 0,1 0,15 0,2 0,25 0 20 40 60 80 100 120 140 160 180 200 H ea rt r at e va ri ab ili ty ( s) Time (s)
In the HRV plot of the high stress disadvantageous group, a decrease in HRV is seen during the moment of choosing (t = 97 – t = 101). Other negative peaks are seen around the time where the choice is introduced (t = 90), at the beginning of the outcome video (t = 138) and at the actual moment of the outcome being told (t = 156). The HRV plot for the low stress, disadvantageous group shows multiple fluctuations and peaks at different moments in the task. In this plot, the beginning of the outcome video can be seen at t = 138. No stress was seen at the decision-making moment. All other peaks in the plot could not be traced back to a certain, stressful moment in the task.
In the HRV plots for the low stress advantageous and the high stress advantageous group, no remarkable changes in HRV are seen during the moment of decision-making, as HRV fluctuates around 0 or increases.
4. Conclusions and discussion
The main goal of this pilot study was to investigate the effect of stress on advantageous and disadvantageous decision-making in the prefrontal cortex. Significant differences in both / and / ratio between the low stress disadvantageous and high stress disadvantageous groups show an effect of stress on disadvantageous decision making. Stress causes a significant difference in / and / ratio increase for people who make disadvantageous choices, as hypothesized. No significant differences were found between the low stress advantageous and the high stress advantageous groups, which means this difference is only present in disadvantageous decision-makers. A significant difference between the low stress advantageous and the low stress disadvantageous group in / ratio differences show that / ratio decreases more in participants that made a disadvantageous choice compared to participants who made an advantageous choice under low stress conditions.
Theta and alpha power differences between the introduction video and the moment of decision-making were principally used as marker for stress. Previous literature showed that alpha suppression and theta increase are typical markers for stress (Subhani et al., 2011). In this pilot study, alpha suppression was found in both low stress groups and in the low stress advantageous group, but an alpha increase was found in the high stress disadvantageous group. Theta decrease was found in all groups, but least in the high stress groups. Markers as described in previous literature were therefore not found. Nevertheless, significant differences between alpha and theta powers were found between the introduction video (rest) and moment of decision-making. This suggests that stress causes a reduced suppression of theta and alpha waves in the prefrontal cortex when disadvantageous decisions are made.
HRV was used to identify differences in stress between all four groups. The average in HRV of the low stress advantageous and the high stress advantageous groups fluctuates around 0, which means HRV did not deviate a lot during the task in comparison to the HRV of the low stress disadvantageous and high stress disadvantageous group. This could mean no stress was induced within these two groups of participants. As for the low stress disadvantageous group, HRV fluctuated over time, but not specifically in the time window for decision-making. This means there was no noticeable stress for participants in this group in the moment of choice, as expected for the low stress group. However, HRV of the low stress disadvantageous group was calculated over only three participants. In the high stress disadvantageous group, a clear decrease in HRV is seen at the moment of choosing. This suggests participants were actually stressed at the moment of choosing. Other peaks reveal HRV changes at other moments of the task, but not all peaks can be traced back to a specific stressful moment in the task. Considering the differences in stress level and random stress
peaks in the HRV plots at different moments in the task, stress has not been induced as desired in both high stress groups and suppressed in both low stress groups. Therefore, time-constraints can be used to induce stress before the moment of decision-making (Svenson & Maule, 1993). Time-constrains also replicate stress experienced by train passengers.
In addition, to improve the experiment, more participants are needed for the validity of the findings. The low stress disadvantageous group only contained data of three participants. The experimental condition with most participants only contained 13 participants. To include more participants for data analysis, reliable markers have to be used to mark the start of the task and for an indication of the decision-making moment.
5. Research proposal
§5.1 Introduction
The findings of the NS stationslab show a significant and relevant effect of stress on the / and / ratio in advantageous and disadvantageous decision-making. Also, alpha and theta frequency differences were found as an effect of stress on disadvantageous decision making. A new study design is required to ensure higher validity of these findings. This research will contribute to a better understanding of stress in the human brain and eventually prediction, prevention and reduction of stress.
5.2 Approach 5.2.1 The task
The experiment will consist of a mental task in which participants will be asked to imagine themselves in a virtual sketched out scenario to investigate the effect of mental stress on decision making. Prior to the mental task, participants need to fill in a questionnaire for general information and additional information which could affect the outcome of the experiment. The general information includes gender, age, weight and length. Additional information includes heart diseases, vascular disorders, mental conditions, stress level, colour-blindness and drug usage. Prior to the task, heart rate and blood pressure of all participants will be measured to create a baseline measurement. EEG and ECG recordings will take place during the
whole mental task. Based upon earlier research of Starcke and colleagues (2012), EEG electrodes will be placed to measure the prefrontal cortex and the parietal cortex. A ground electrode will be placed on Cz and a reference will be placed upon the right ear. Figure 8 shows a visual representation of the electrodes in an EEG cap. In addition, ECG electrodes will be placed upon the left wrist of the participant to measure heart beat and finally calculate HRV. The task will start with a baseline EEG and ECG measurement which consists of sitting still for 20 seconds. Participants will be asked to look at a fixation point during this baseline measurement. Next, an introduction video will be shown to illustrate a situation in which participants are expected to envision themselves. In this video, all participants are told to have
Figure 8: visual representation of electrode placement
an important job interview they go to by train. Next, a video will show the participants how much time there is left before the train leaves: 2 minutes (high stress), 8 minutes (medium stress) or 25 minutes (low stress). Participants will be randomly distributed over these groups. Based upon the given time participants have before their train leaves, participants are told how much time they have to make a choice between running to their train, walking to their train or visit some shops before going to the train. If participants have 2 minutes left, they are asked to make their decision within 2 seconds to induce time-based stress. If participants have 8 minutes left, they are asked to make their decision within 4 seconds to induce time-based stress. If participants have 25 minutes left, they are asked to make their decision within 25 seconds, so there is no induction of time-based stress. Induction of time-based stress in this experiment will replicate stress of train passengers in real life better than any other form of stress induction.
After participants know how much time they have left, they are being told what choices they are able to make: run to their train, walk to their train or go shopping before they go to their train. Choices will only be explained after there is told how much time there is left to make the choice to prevent participants from making their choice prior to the moment of decision-making and causing a defect in the induction of time-based stress.
Next, participants are asked to make their choice within the given time slot, visualized by a time-bar. The choice is made by pressing a button: ‘R’ for running, ‘W’ for walking and ‘S’ for shopping. Thereafter, a video with the outcome of their choice will be shown: whether the participant caught the train or not. Figure 9 shows a visual representation of the experimental task. Markers will be used to mark the start of the experiment, to mark the start of the moment of decision-making, the moment the choice was made and the ending of the experiment.
Figure 9: visual representation of the experimental task
The outcome of the experiment is based upon the choice participants made and is presented in table 4. Based upon the outcome, participants will be shown a video of someone who caught its train or someone who missed its train.
Table 4: Outcome of the choice participants made based upon choice and given time
Time Choice Outcome
2 minutes Run Caught (50%) or missed (50%) based upon randomness
Walk, shop Missed
8 minutes Run Caught
Walk Caught (50%) or missed (50%) based upon randomness
25 minutes Run, walk Caught
Afterwards, heart rate and blood pressure will be measured. In addition, participants will be asked to fill in an outtake questionnaire about the experimental task. This questionnaire includes mental state and stress level. Additionally, participants are asked whether they completely understood the task, to eventually filter out errors of participants who didn’t understand what choices they could make or participants who pressed a wrong button. Filtering out these errors will prevent that participants will be sort out in the wrong experimental group and that wrong data will be used.
All participant data will be anonymized and ethical guidelines of the University of Amsterdam will be used throughout the experiment.
5.2.2 Data analysis
Based upon the time participants were given to catch the train (2, 8 or 25 minutes) and the choice they made (running, walking or shopping) participants are classified into six different groups as presented in table 5. To get 30 participants in every experimental condition, at least 180 participants are needed for the validity of the experiment. Based upon the NS stationslab pilot study, only 10% of all participants ended up in the high stress disadvantageous group, resulting in a minimum of 300 participants needed.
Table 5: subgroups of the experimental task High stress advantageous High stress disadvantageous Medium stress advantageous Medium stress disadvantageous Low stress advantageous Low stress disadvantageous
Time 2 minutes 2 minutes 8 minutes 8 minutes 25
minutes
25 minutes
Choice Run Walk, shop Run, walk Shop Walk,
shop
Run
For all participants, delta (1-3 Hz), theta (4 – 7 Hz), alpha (8 – 12 Hz) and beta (13 – 30 Hz) frequency powers will be determined during the baseline measurement and the moment of decision-making. In-group differences will reveal the effect of stress on increase or decrease of one certain frequency band. Between group differences will reveal how big the effect of stress is on decision-making. / and / ratio differences reveal the effect of stress on advantageous and disadvantageous decision-making.
5.2.3 Work plan
Participant data will be collected by the NS-stationslab for a variety of participants who travel past ‘Amsterdam Centraal’ train station. The research will take approximately 10.5 months. Which will allow students to contribute as research assistant for one or two semesters. An overview of the work plan is seen in figure 10.
5.3 Scientific and societal impact
This research will bring scientific insights about the effect of stress on decision making, which has to my knowledge never been researched before. It will reveal the operation of brain areas involved in stressful decision-making and therefore a better scientific understanding of stress in the human brain and how people act under stressful conditions.
In addition, this research will bring a better understanding of stress-based decisions and can therefore contribute to the knowledge of predicting, preventing and reducing stress. Data of the NS stationslab can be used to improve overall train station experiences and respond to choices of train passengers.
6. References
Brand, M., Labudda, K., & Markowitsch, H. J. (2006). Neuropsychological correlates of decision-making in ambiguous and risky situations. Neural Networks, 19(8), 1266-1276.
De Bellis, M. D., Keshavan, M. S., Clark, D. B., Casey, B. J., Giedd, J. N., Boring, A. M., ... & Ryan, N. D. (1999). AE Bennett Research Award: developmental traumatology: part II. Brain development. Biol
Psychiatry, 45(10), 1271-84.
Goldstein, D. S. (1995). Stress, catecholamines, and cardiovascular disease. Oxford University Press. Heuser, I., & Lammers, C. H. (2003). Stress and the brain. Neurobiology of aging, 24, S69-S76. De Kloet, E. R., Joëls, M., & Holsboer, F. (2005). Stress and the brain: from adaptation to disease. Nature reviews neuroscience, 6(6), 463-475.
Harmony, T., Fernández, T., Silva, J., Bernal, J., Díaz-Comas, L., Reyes, A., ... & Rodríguez, M. (1996). EEG delta activity: an indicator of attention to internal processing during performance of mental tasks. International journal of psychophysiology, 24(1-2), 161-171.
Levine, S. (2000). Influence of psychological variables on the activity of the hypothalamic–pituitary– adrenal axis. European journal of pharmacology, 405(1-3), 149-160.
McEwen, B. S., & Sapolsky, R. M. (1995). Stress and cognitive function. Current opinion in
neurobiology, 5(2), 205-216.
Mendl, M. (1999). Performing under pressure: stress and cognitive function. Applied Animal Behaviour
Science, 65(3), 221-244.
Munck, A., Guyre, P. M., & Holbrook, N. J. (1984). Physiological functions of glucocorticoids in stress and their relation to pharmacological actions. Endocrine reviews, 5(1), 25-44.
Saeed, S. M. U., Anwar, S. M., Majid, M., & Bhatti, A. M. (2015, December). Psychological stress measurement using low cost single channel EEG headset. In 2015 IEEE International Symposium on
Signal Processing and Information Technology (ISSPIT) (pp. 581-585). IEEE.
Schutter, D. J., & Van Honk, J. (2005). Electrophysiological ratio markers for the balance between reward and punishment. Cognitive Brain Research, 24(3), 685-690.
Staal, M. A. (2004). Stress, cognition, and human performance: A literature review and conceptual framework.
Starcke, K., & Brand, M. (2012). Decision making under stress: a selective review. Neuroscience &
Biobehavioral Reviews, 36(4), 1228-1248.
Subhani, A. R., Xia, L., & Malik, A. S. (2011). EEG signals to measure mental stress. In 2nd
International Conference on Behavioral, Cognitive and Psychological Sciences (pp. 84-88). Maldives.
Svenson, O., & Maule, A. J. (Eds.). (1993). Time pressure and stress in human judgment and decision
making. Springer Science & Business Media.
Taelman, J., Vandeput, S., Spaepen, A., & Van Huffel, S. (2009). Influence of mental stress on heart rate and heart rate variability. In 4th European conference of the international federation for medical
and biological engineering (pp. 1366-1369). Springer, Berlin, Heidelberg.
van Ravenswaaij-Arts, C. M., Kollee, L. A., Hopman, J. C., Stoelinga, G. B., & van Geijn, H. P. (1993). Heart rate variability. Annals of internal medicine, 118(6), 436-447.
