Mind-Body Problem: Investigating the Gap Between Physiological and Perceived Stress Using Smartwatch Data

Petri Salo MIND-BODY PROBLEM: INVESTIGATING THE GAP BETWEEN PHYSIOLOGICAL AND PERCEIVED STRESS USING SMARTWATCH DATA JYVÄSKYLÄN YLIOPISTO INFORMAATIOTEKNOLOGIAN TIEDEKUNTA 2024 Salo, Petri Mind-Body Problem: Investigating the Gap Between Physiological and Perceived Stress Using Smartwatch Data Jyväskylä: University of Jyväskylä, 2024, 77 pp. Cognitive Science, Master’s Thesis Supervisor: Kujala, Tuomo The popularity of physiological sensors is rising, with more people using smartwatches to monitor stress and optimize performance based on physiological data. However, recent studies show that physiological stress often differs from perceived stress, which can lead users to misinterpret these signals and make detrimental health decisions. This study investigated factors contributing to this discrepancy and the mechanisms underlying perceived stress. A multilevel linear mixed-effects model was applied to examine the relationship between perceived and physiological stress and to evaluate the moderating effects of attentional styles, the Big Five personality traits, and metacognitive beliefs on this relationship. Self-report questionnaires were used to assess these psychological factors and perceived stress. Physiological stress was measured daily for one month using Garmin smartwatches in a sample (N = 5) of highly educated adults aged 28–45. Results indicated no association between physiological and perceived stress nor moderation by attentional styles, personality traits, or metacognitive beliefs. Exploratory findings suggested that conscientiousness, extraversion, and metacognitive beliefs predicted perceived stress. Understanding how users interpret and trust physiological stress measurements is vital for developing user-friendly technologies that support informed decision-making and promote well-being. Despite its limitations, this study proposes a methodology for identifying factors influencing the relationship between physiological and perceived stress. Keywords: attentional styles, the Big Five personality traits, linear mixed-effects model, metacognitive beliefs, perceived stress, physiological measurements, smartwatches ABSTRACT Salo, Petri Mieli–ruumis-ongelma: Fysiologisen ja koetun stressin välisen eron tutkiminen älykellodatan avulla Jyväskylä: Jyväskylän yliopisto, 2024, 77 s. Kognitiotiede, pro gradu -tutkielma Ohjaaja: Kujala, Tuomo Fysiologisten sensorien suosion kasvaessa yhä useammat ihmiset käyttävät älykelloja seuratakseen stressiä ja optimoidakseen suorituskykyään fysiologisen datan perusteella. Tutkimukset osoittavat, että fysiologinen stressi eroaa usein koetusta stressistä, mikä voi saada käyttäjät tulkitsemaan näitä signaaleja väärin ja tekemään haitallisia terveyteen liittyviä päätöksiä. Tässä tutkimuksessa tutkittiin tähän ristiriitaan vaikuttavia tekijöitä ja koetun stressin taustalla vaikuttavia mekanismeja. Lineaarista monitasomallia käytettiin koetun ja fysiologisen stressin välisen suhteen tutkimisessa sekä arvioidessa tarkkaavaisuustyylien, viiden suuren persoonallisuuspiirteen ja metakognitiivisten uskomusten moderoivia vaikutuksia tähän suhteeseen. Näitä psykologisia tekijöitä ja koettua stressiä arvioitiin kyselylomakkeilla. Fysiologista stressiä mitattiin päivittäin yhden kuukauden ajan Garminälykelloilla korkeasti koulutetuilla 28–45-vuotiailla aikuisilla (N = 5). Tuloksissa ei havaittu yhteyttä fysiologisen ja koetun stressin välillä, eivätkä tarkkaavaisuustyylit, persoonallisuuden piirteet tai metakognitiiviset uskomukset moderoineet yhteyttä. Eksploratiiviset tulokset viittasivat siihen, että tunnollisuus, ekstraversio ja metakognitiiviset uskomukset ennustivat koettua stressiä. Ymmärrys käyttäjien fysiologisiin stressimittauksiin kohdistuvasta luottamuksesta ja mittausten tulkinnasta on tärkeää kehitettäessä käyttäjäystävällisiä teknologioita, jotka tukevat tietoista päätöksentekoa ja edistävät hyvinvointia. Rajoitteistaan huolimatta tutkimus tarjoaa metodologisia ehdotuksia fysiologisen ja koetun stressin suhteeseen vaikuttavien tekijöiden tunnistamiseksi. Asiasanat: fysiologiset mittaukset, koettu stressi, lineaarinen monitasomalli, metakognitiiviset uskomukset, tarkkaavaisuustyylit, viisi suurta persoonallisuuden piirrettä, älykellot TIIVISTELMÄ FIGURES FIGURE 1 The Hypothesis 1. ..................................................................................... 12 FIGURE 2 The Hypotheses 2, 3, and 4. ..................................................................... 16 FIGURE 3 Distribution of the Residuals for Model 1. ............................................ 29 FIGURE 4 Normality Q-Q Plot of the Residuals for Model 1. .............................. 30 FIGURE 5 Examination of the Homoscedasticity for Model 1. ............................ 30 FIGURE 6 Distribution of the Residuals for Model 2. ............................................ 33 FIGURE 7 Normality Q-Q Plot of the Residuals for Model 2. .............................. 34 FIGURE 8 Examination of the Homoscedasticity for Model 2. ............................ 34 TABLES TABLE 1 Descriptions of the Big Five Personality Dimensions. .......................... 12 TABLE 2 Descriptive Statistics of the Measures. .................................................... 26 TABLE 3 The multilevel linear mixed-effects model of perceived stress. ........... 27 TABLE 4 The multilevel linear mixed effects model of perceived stress with weekly averages. .......................................................................................................... 32 TABLE OF CONTENTS ABSTRACT TIIVISTELMÄ FIGURES AND TABLES 1 INTRODUCTION ................................................................................................. 7 2 CONCEPTUALIZING STRESS: PERCEPTION, PHYSIOLOGY, AND CORRELATES ............................................................................................................... 10 2.1 Perceived stress and physiological stress .................................................. 10 2.2 Personality ...................................................................................................... 12 2.3 Attentional styles ........................................................................................... 13 2.4 Metacognitive beliefs .................................................................................... 15 3 METHOD ............................................................................................................. 17 3.1 Hypotheses ..................................................................................................... 17 3.2 Participants ..................................................................................................... 17 3.3 Procedure ....................................................................................................... 18 3.4 Measures ......................................................................................................... 20 Attentional Style Questionnaire .............................................................. 20 Big Five Inventory Short-Form ................................................................ 21 The Perceived Stress Scale ........................................................................ 21 The Metacognitions Questionnaire ......................................................... 22 Smartwatches ............................................................................................. 22 3.5 Statistical analysis ......................................................................................... 24 4 RESULTS .............................................................................................................. 26 4.1 The null model ............................................................................................... 26 4.2 Model 1, Hypothesis 1: Perceived vs. physiological stress in multilevel model ........................................................................................................... 27 4.3 Residuals and homoscedasticity of Model 1 ............................................. 28 4.4 Hypotheses 2–4: Personality, attentional styles, and metacognitive beliefs as interaction terms ....................................................................... 31 4.5 Model 2: Predicting perceived stress with weekly physiological stress averages ....................................................................................................... 31 4.6 Residuals and homoscedasticity of Model 2 ............................................. 32 4.7 Personality, attentional styles, and metacognitive beliefs as interaction terms in Model 2 ........................................................................................ 35 5 DISCUSSION ....................................................................................................... 36 5.1 Limitations ..................................................................................................... 41 5.2 Future directions ........................................................................................... 43 6 CONCLUSION .................................................................................................... 45 REFERENCES ................................................................................................................ 46 APPENDIX 1 INCLUSION/EXCLUSION CRITERIA FORM ............................... 63 APPENDIX 2 BASELINE QUESTIONNAIRE .......................................................... 65 APPENDIX 3 ATTENTIONAL STYLE QUESTIONNAIRE ................................... 72 APPENDIX 4 BIG FIVE INVENTORY SHORT-FORM .......................................... 74 APPENDIX 5 THE METACOGNITIONS QUESTIONNAIRE .............................. 75 APPENDIX 6 THE PERCEIVED STRESS SCALE .................................................... 77 With the growing popularity of wearable electronics, more people are using their devices to monitor their health regularly. Accordingly, Dhingra et al. (2023) reported that over 25% of US adults routinely used smart wearables in 2019 and 2020. Similarly, Vogels (2020) stated that nearly 20 % of Americans used a smartwatch or fitness tracker. The ownership of wearable devices is rapidly becoming more popular since Kunst (2023) reported that 41 % of Americans own a smartwatch or a fitness tracker. According to the statistics, the ownership varies between 41 % and 55 % in the UK, China, India, Germany, and Canada (Kunst, 2023). Thus, every second person worldwide is expected to own a smartwatch or a fitness tracker, ushering in a new era of health tracking. These gadgets are not only popular for their stylish, bracelet-like design, but they can also produce accurate data, allowing users to understand their bodily signals (Deng et al., 2023). In many circumstances, observing physiological signals can be advantageous to individuals. An older adult may recognize indicators of underlying or present health issues by monitoring blood oxygen concentration or heart rate variation using smart wearables. An office worker can optimize their efficiency by tracking the quality of their sleep and recovery after office hours with the help of the figures. By tracking fluctuations in oxygen uptake capacity, both regular and competitive athletes can monitor their endurance progress and modify the intensity of their workouts according to the device's indicated recovery level. Whether a daily exerciser, a pensioner, a top athlete, or an office worker, the recuperation and, consequently, the performance of any group can be affected to some extent by the amount of stress experienced during the day. To monitor stress levels daily, some manufacturers have developed technologies for their smart wearables to calculate stress metrics (Li et al., 2023). However, there is debate about their ability to estimate stress levels. Fazeli et al. (2022) have declared that “There is growing evidence that there are notable differences in how individuals perceive their stress levels and the actual physiological manifestation of a stress response” (p. 2894). One meta-analysis revealed a significant association between physiological and 1 INTRODUCTION 8 perceived emotional stress in only 25% of the social stress studies examined (Campbell & Ehlert, 2012, as cited in Fazeli et al., 2022). Hence, the manufacturerspecific calculation formula alone might not be an accurate method for measuring stress experiences. Nonetheless, should data from these devices be interpreted by every second individual in the future, assessing the validity and reliability of their stress estimates will be crucial, especially given the potential impact on users such as elite athletes. Reliability refers to the consistency of these estimates over time, while validity indicates how accurately they reflect actual stress levels. Misinterpreting physiological data or overlooking accumulated stress could increase the risk of injury or overload in athletes, potentially leading to inadequate recovery or inappropriate training intensity (Mónico et al., 2020; Pensgaard et al., 2018). Research by Ristolainen et al. (2012) shows that injuries can be critical factors in the continuity and success of an elite athlete's career. Therefore, if smart wearables play a meaningful role in optimizing an athlete's training, studying the factors influencing the relationship between their stress estimates and perceived stress is critical. This would help identify the limitations of these devices in assessing stress and guide users in interpreting the data. The validity and reliability of device measurements for stress are crucial factors that significantly influence user experience (Maher et al., 2017). In surveys targeting smartwatch users (Lu & Zhou, 2024), individuals have reported making concrete health-related decisions based on their devices' measurements and feedback. If users continuously question the devices’ ability to estimate stress accurately and make unsatisfactory decisions based on the data, they may eventually stop using them. To address this, these factors could be considered when smartwatches collect user data, allowing the presented information to be customized according to each individual’s psychological and cognitive characteristics. Specifically, informing users about the factors influencing the relationship between physiological and perceived stress would empower them to assess the measurements critically, make informed decisions, and potentially improve their overall experience with the device. By far, current research has not investigated which psychological and cognitive factors might explain the discrepancy between users’ stress estimates and the stress scores provided by smart wearables. Motivated by this gap, our study investigates the strength of the relationship between perceived and physiologically measured stress, examining whether cognitive and psychological factors—such as personality, metacognitive beliefs, and attentional styles—moderate this relationship. In past studies, personality traits have been linked to stress reactivity (Bolger & Schilling, 1991; Bolger & Zuckerman, 1995; Xin et al., 2017), and metacognitive beliefs have correlated to perceived stress levels (Ramos- Cejudo & Salguero, 2017; Sariçam, 2015; Spada et al., 2008). Although the potential link has not yet been explored for attentional styles, a relationship with perceived stress has been found in studies with measures containing the same items as the attentional styles measure used in this study (Hepburn et al., 2021). 9 The current study aims to identify the factors contributing to the gap between physiological and perceived stress by examining their relationships. Discussing these individual factors is necessary to better understand the potential relationships. 16 Individuals with such metacognitive beliefs may experience an increased focus on their thoughts, perceiving them as threatening. Therefore, we are interested in how unhelpful metacognitive beliefs may affect stress perception and their relationship with stress levels measured by smartwatches. Studies have shown that perceived stress levels positively correlate with unhelpful metacognitive beliefs (Ramos-Cejudo & Salguero, 2017; Sariçam, 2015; Spada et al., 2008). Consequently, individuals with higher levels of unhelpful metacognition are expected to report higher perceived stress scores. We hypothesize that metacognitive beliefs moderate the relationship between perceived and physiologically measured stress. This expectation arises because an increased focus on one’s thoughts and cognition is likely to induce stress, particularly when an individual feels their knowledge or skills are inadequate for a given situation. Additionally, those overly focused on their thoughts may be more attuned to sympathetic nervous system activation, such as an increased heart rate, interpreting these physiological responses as stress. Conversely, individuals with lower metacognitive beliefs tend not to dwell on personal cognition or stressful thoughts and may experience lower stress perceptions than those who do. In addition to examining how levels of metacognitive beliefs indicate an individual’s focus on cognition and negative thinking, we aim to investigate whether these cognitive processes moderate the relationship between stress perception and physiological stress. This hypothesis is illustrated in Figure 2. The methods section will provide a detailed discussion of how we will test the assumptions. FIGURE 2 The Hypotheses 2, 3, and 4. H2 H3 H4 Physiological stress (X) Perceived stress (Y) Metacognitive beliefs (M) Personality (M) Attentional styles (M) 17 3.1 Hypotheses This study aims to reveal why physiological stress estimates differ from perceived stress experiences. To test this, we have formulated the following hypotheses: H1: A positive correlation exists between physiological and perceived stress. H2: There is a moderating effect of personality on the relationship between physiological and perceived stress. H3: Attentional styles have a moderating effect on the relationship between physiological and perceived stress. H4: There is a moderating effect of metacognitive beliefs on the relationship between physiological and perceived stress. 3.2 Participants We used both online and on-site advertisements to recruit participants to our study. Initially, we analyzed local sports clubs to determine their participation eligibility and potential for the current study. We created a list of all sports clubs and channels in Jyväskylä and sorted them based on their size, sports category, availability of contact information, and members’ average sports profile. Subsequently, we ranked them based on their potential study 3 METHOD 18 participation. We used the size of the club and the availability of the contact information as criteria for the ranking. We created posters to recruit participants, targeting three groups: athletes, coaches, and sports enthusiasts. The posters provided brief information on the study procedure, the principal investigator's contact details, a link, and a QR code directing to the study website (https://stress-study.it.jyu.fi/), where participants could sign up. We distributed the posters via email to our contacts, who shared them through the internal channels of their sports clubs. The study was also promoted on the University of Jyväskylä's platforms, particularly in the Faculties of Information Technology and Sport and Health Sciences. Additionally, we printed copies of the posters and displayed them in the university facilities and local sports centers. Participants initially responded to inclusion and exclusion criteria questions via a link in the advertisement materials. To participate, individuals were required to be at least 18 years old, own a Garmin smartwatch, and have no prior diagnoses of sleep disorders, anxiety disorders, posttraumatic stress disorder, or regular use of certain medications (e.g., painkillers, blood pressure medications, sedatives). If participants did not meet these criteria, they were informed that they could not participate due to data generalizability issues. For participants who met the criteria, we emailed individual links to fill out the first questionnaire. We provided them with instructions on how to connect their smartwatches to the UTV platform (https://utv.it.jyu.fi/). Participants were informed about the voluntary nature of the study and the confidentiality of the data collection. Additionally, all respondents filled out an informed consent form. Consequently, our final sample consisted of 5 participants, of whom 80% were males. The mean age of the respondents was 34, with a range of 28 and 45. The average level of education of the respondents (highest completed degree) was a master’s degree or equivalent. In contrast, one of the participants had completed a bachelor's degree, and the other had a doctoral degree. 3.3 Procedure The study was approved by The Human Sciences Ethics Committee of the University of Jyväskylä (1675/13.00.04.00/2023). This research was part of a larger study that involved a six-month online data collection using the webbased UTV platform. The following questionnaires were included in the study: Attentional Style Questionnaire (ASQ), Big Five Inventory Short-Form (BFI-10), Epworth Sleepiness Scale (ESS), General Self-efficacy Scale (GSE), The 22-item Ruminative Response Scale (RRS-22), The Godin-Shephard Leisure-Time Physical Activity Questionnaire, The Metacognitions Questionnaire (MCQ-30), and The Perceived Stress Scale (PSS) (Amireault & Godin, 2015; Cohen et al., 1983; Johns, 1991; Rammstedt & John, 2007; Schwarzer & Jerusalem, 1995; Treynor et al., 2003; Van Calster et al., 2018; Wells & Cartwright-Hatton, 2004). 19 The questionnaires were converted into HTML format and integrated into the UTV platform. Since the questionnaires were in English and our participants were Finnish, we aimed to ensure the understandability of complex phrases, given their varying English proficiency. Following Head et al. (2021), we added tooltip functionality to the HTML forms, displaying translations of selected words or phrases when hovered over with the cursor. In our forms, the tooltip functionality worked similarly, as Head et al. (2021) described in their paper: “When a reader comes across a nonce word that they do not understand, ScholarPhi lets them click the word to view a position-sensitive definition in a compact tooltip” (p. 1). We collected the data in two parts: by using a baseline survey to collect demographics and less dynamic concepts and a weekly questionnaire to assess dynamic variables. The data collection started with an inclusion/exclusion form, which participants filled out once they followed the registration link on the study website. The inclusion/exclusion form is presented in Appendix 1. After they had completed the form, eligible participants received links to the baseline survey. The baseline survey consisted of questions related to demographics, sports profiles, and three questionnaires: ASQ, BFI-10, and MCQ-30 (Rammstedt & John, 2007; Van Calster et al., 2018; Wells & Cartwright-Hatton, 2004). The items of the baseline survey are presented in Appendices 2, 3, 4, and 5. The registration process, including the inclusion/exclusion form, baseline survey, and smartwatch connection, took about 20 minutes. The first measurement point was the baseline survey, followed by weekly measurements over the next six months, as detailed below. The participants received email reminders every Monday at 8:30 AM to fill out the weekly questionnaire in the UTV platform, which comprised a set of validated questionnaires. Daily reminders to the participants were sent at 7:30 PM within the week until they had answered the questionnaire. The included questionnaires were The Godin-Shephard Leisure-Time Physical Activity Questionnaire, GSE, ESS, PSS-10, ASQ, and RRS-22 (Amireault & Godin, 2015; Cohen et al., 1983; Johns, 1991; Schwarzer & Jerusalem, 1995; Treynor et al., 2003). The participants spent approximately 8 to 10 minutes to complete the weekly questionnaire. The weekly questionnaire is presented in Appendix 6. While this weekly data was collected in the context of the larger study, it was not utilized in this study apart from the Perceived Stress Scale measurements. Additionally, we collected physiological data from participants’ Garmin, Polar, and Suunto smartwatches through the UTV platform. Depending on the device, this data included stress level, heart rate (HR), heart rate variability (HRV), location, intensity minutes, movement (in steps and distance), calories, sleep duration, and phases, as well as workout session data. To enable data collection, participants selected their smartwatch manufacturer from a list, logged into the manufacturer's service, and were granted permission to transfer their physiological data from the cloud service to the UTV platform. Consequently, the data was transferred to the UTV automatically and continuously during the study whenever participants synced data from their devices to the 20 manufacturer’s cloud service. As this data contained identifiable personal data, we took appropriate measures to convert the data into an unidentifiable format. Our data contained personal information on circadian rhythms, daily activities, and location data. This urged us to transform the data into a pseudonymous format. Thus, when participants registered for the study, we created their accounts using pseudonymous user identities, which prevented us from identifying individual participants. Furthermore, we calculated intervals instead of individual values, such as weight classes instead of the participant's weight. This was sufficient because we wanted to investigate whether there were significant effects at the group level. Despite pseudonymization, it was still possible for registered data processors to convert pseudonymous personal data to identifiable data, but it was not done without an appropriate reason. However, there are always risks involved in personal data processing, and we will discuss these risks and our efforts to minimize them. Personal data is subject to information security risks such as data breaches, unauthorized access, and accidents in data processing. To minimize the risk of data breaches, we conducted weekly removal of the data in our collection database, and that data was encrypted and stored in our research database, which was only accessible from dedicated machines from the university network. To avoid unauthorized access to personal data, we registered personal data processors for the study, and these individuals signed a data processing agreement. Additionally, we prevented unauthorized access to personal data with a password and username, which were only given to registered data processors, and all logins to the database were registered to monitor usage. To further minimize the information security risks involved, we agreed to remove all identifiable personal data upon completion of the study. We informed the participants of the information security procedure by describing the procedure in the informed consent form signed by and collected from each participant. 3.4 Measures Attentional Style Questionnaire (ASQ; Van Calster et al., 2018) The Attentional Style Questionnaire consists of 17 items, measuring both internal and external dimensions of attention along with top-down versus bottom-up based attention. Items 2, 7, 8, 9, 10, 12, 13, and 15 measure internal attention, while items 1, 3, 4, 5, 6, 11, 14, 16, and 17 are external attention oriented. Items 4, 5, 6, 10, 11, 14, and 15 are reverse-scored. The items are measured with a 6-point Likert scale, from 1, “totally disagree,” to 6, “totally agree.” An example of a bottom-up-oriented item is “I have trouble concentrating when there is movement in the room I am in”. Alongside, the following item represents top-down oriented attention: “In general, I stay in control of my thoughts and do not let myself get distracted by interfering thoughts”. A higher score refers to a bottom- 21 up oriented attentional style, while a lower score indicates that a person tends to engage in top-down attentional processes. Based on the scores, attentional styles are classified into four different categories: mostly distracted by external stimuli, mostly distracted by internal stimuli, not easily distracted, and easily distracted. To estimate the internal reliability of the scale, we used Cronbach’s alpha, for which values > .70 are acceptable (Bland & Altman, 1997). Van Calster et al. (2018) have reported the following values for the scale with their sample: for the first factor, internal attention, internal consistency (Cronbach’s alpha) was .82. By contrast, for external attention, the reliability was .81. In our study, the internal consistency for items measuring internal attention was .85, and .89 for external attention. To ensure the best understanding of the questions in our sample, we used the tooltip functionality for question 4, “I can be so absorbed by a line of thoughts that I become more or less unaware of my surroundings”, question 12 “Sometimes I interrupt an activity to check an unrelated detail”, and question 15 “I can spend several minutes on a question and try to dissect it”. Big Five Inventory Short-Form (BFI-10; Rammstedt & John, 2007) The Big Five Inventory Short-Form is a 10-item self-report scale that measures five dimensions of personality: neuroticism, openness to experiences, extroversion-introversion, conscientiousness, and agreeableness. The items are measured within a 5-point Likert scale, in which one (1) stands for “Disagree strongly” and five (5) for “Agree strongly”. Regarding internal consistency (Cronbach’s alpha), Rammstedt and John (2007) reported sufficient levels of consistency across the subscales: extraversion .89, agreeableness .74, conscientiousness .82, neuroticism .86, and openness .79. The corresponding values in our study were .32, .22, .93, .48, and .90, respectively. According to Cortina (1993), Cronbach's alpha values decrease with fewer items in the instrument. Thus, the rule of > .70 cannot be directly applied to two-item subscales. As explained in the study procedure, we used the tooltip functionality in question 5, “I see myself as someone who has few artistic interests,” to ensure the Finnish sample understood this question. The Perceived Stress Scale (PSS-10; Cohen et al., 1983; Cohen & Williamson, 1988) The Perceived Stress Scale combines ten items concerning feelings and thoughts during the last month. Each answer will be given within a 5-point Likert scale concerning rumination frequency, from 0 “never” to 4 “very often”. Example items of the scale are: “In the last month, how often have you felt that you were on top of things?” or “In the last month, how often have you been angered because of things that were outside your control?”. Higher scores reflect higher levels of perceived stress. The scores are classified as follows: 0-13 points reflect low stress, 14-26 points moderate stress, and 27-40 points represent a high stress score. The scale takes 5-10 minutes to complete. Considering the reliability scores of the PSS-10, Cohen and Williamson (1983) reported its internal consistency (Cronbach’s alpha) as .78. With our sample, the reliability was .87. 22 In this questionnaire, we implemented the tooltip feature for the following questions: 8. “In the last month, how often have you felt that you were on top of things?” and 10. “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?”. The Metacognitions Questionnaire (MCQ-30; Wells & Cartwright-Hatton, 2004) The MCQ-30 is a self-report scale with 30 items, measured with five subscales. The five subscales of the MCQ-30 are cognitive confidence, positive beliefs about worry, cognitive self-consciousness, negative beliefs about the uncontrollability of thoughts and danger, and beliefs about the need to control thoughts (Wells & Cartwright-Hatton, 2004). Metacognitive beliefs are reported in a 4-point Likert scale, in which point one (1) means “completely agree” and four (4) stands for “do not agree”. The total scores range from 30 to 120, with higher scores reflecting increased levels of unhelpful metacognitive beliefs. Depending on the purpose, the instrument can be used with subscales or as a total scale. Within the original scale, the internal consistency (Cronbach’s alpha) for the total scale was .92. For each subscale, the reliability scores were .93, .92, .92, .91, and .71 for cognitive confidence, positive beliefs, cognitive self-consciousness, uncontrollability, and danger, and need to control thoughts, respectively (Wells & Cartwright-Hatton, 2004). In our sample, the Cronbach alpha reliability was .67 for the total scale and .92, .97, .71, .95, and .76 for the subscales. In this scale, we implemented the tooltip feature for questions 12 “I monitor my thoughts” and 30 “I constantly examine my thoughts,” as we expected that understanding them could be problematic. Smartwatches (Garmin) The following Garmin smartwatches were utilized in the study: Garmin Vivoactive 4, Garmin Forerunner 965, Garmin Fenix 7 series, Garmin Fenix 6 series, and Garmin Approach S62. Garmin smartwatches utilize wrist-based heart rate measurements using photoplethysmography (PPG) (Hehlmann et al., 2021). PPG technology mainly uses photodetectors and light-emitting diodes(LEDs) (Lee et al., 2021). When the user wears the watch, the green LEDs on the back of the watch illuminate the skin tissue of the wrist, and photodetectors detect the light reflected, indicating volumetric changes in blood (Allen, 2007; Hehlmann et al., 2021; Lee et al., 2021). Consequently, these changes in blood flow are used to estimate heart rate. Garmin smartwatches calculate a stress value based on heart rate measurements, ranging from 0 to 100 (Jerath et al., 2023). These values are categorized as high stress (76-100), medium stress (51-75), low stress (26-50), and no stress or rest (0-25) (Hehlmann et al., 2021). The stress score is calculated using an algorithm from Firstbeat Analytics Ltd., which analyzes heart rate variability (Kettunen & Saalasti, 2005; Firstbeat Technologies Ltd., 2014; Garmin Support, 2020; Hehlmann et al., 2021). The calculation process from heart rate to stress score is detailed in the patent document by Kettunen and Saalasti (2005). First, the recorded heart 23 period signal, such as the time between successive heartbeats, is converted into an equidistant heart rate time series. Then, statistical methods separate segments with similar physiological properties. These segments are assessed to identify cardiological activity unrelated to stress, such as rapid heart rate fluctuations caused by movement or physical load. Next, the remaining heart rate segments are identified as potentially containing stress states. This process uses both low- and high-frequency components of heart rate variability. The low-frequency component reflects sympathetic nervous system activity and potential stress, while the high-fre- quency component suggests parasympathetic nervous system activity, indicating relaxation. The algorithm distinguishes periods of high heart rate with high sympathetic activity and low parasympathetic activity. Finally, the algorithm calculates the stress value by relating heart rate and corrected time (CT) to the spectral powers of heart rate variability (HRV). Spectral power refers to the distribution of power into frequency components, while corrected time accounts for inconsistencies in HRV due to variations in the respiratory cycle or other fluctuations in the respiratory signal (Kettunen & Saalasti, 2005). Although the underlying mechanism of Firstbeat's stress measurement is well-documented, its validity as a physiological stress measure has yet to be confirmed. However, studies have assessed the ability of Garmin smartwatches to measure heart rate. The validity of smartwatch heart rate measurements has been evaluated by comparing them to clinical HR measurement devices, particularly electrocardiograms (ECG), which are considered a reliable standard for measuring HR based on the heart’s electrical activity (Bent et al., 2020; Phillipos et al., 2015). Lin's concordance correlation coefficient (CCC) (Lin, 1989) has been used to statistically compare PPG heart rate measurements to ECG heart rate measurements, indicating how well the two methods agree in measuring the same continuous variable (Akoglu, 2018). Merrigan et al. (2022) compared Garmin Fenix 6 HR measurements to corresponding ECG (Bittium Faros 180™) measurements in multiple physical training tasks to investigate the potential differences. Garmin Fenix 6 measurements showed acceptable agreement with the ECG device during submaximal cycling (CCC: 0.964) and maximal effort ruck (CCC: 0.954) tasks. According to Lin (1989), these CCC values are substantial. Furthermore, the accuracy of PPG measurements in smartwatches depends on the intensity level of the physical task performed while worn (Düking et al., 2020; Hajj-Boutros et al., 2021). When the exercise intensity increases, the accuracy can decrease because of, e.g., rapid wrist movements (Bent et al., 2020; Merrigan et al., 2022). The inaccuracy can be caused by changes in blood flow in the arm or by the excessive movement of the watch on the wrist (Bent et al., 2020). Furthermore, the accuracy of HR measurement can be affected by skin color or tattoos since darker skin colors or dark tattoos do not reflect light equally as lighter skin tones, with the typical green LED (Wallen et 24 al., 2016; Reddy et al., 2018). However, the findings concerning the light reflection of skin are contradictory (Sañudo et al., 2019; Bent et al., 2020; Merrigan et al., 2022). 3.5 Statistical analysis To test our hypotheses of the potential connection between physiological and perceived stress and the potential moderation effects of personality, attentional styles, and metacognitive beliefs on the relationship between physiological and perceived stress, we will use a linear mixed model (LMM). LMM will be used since our data does not meet the criteria of independent observations (Poole & O’Farrell, 1971) of linear regression analysis (LRA). Subsequent stress measurements likely depend on each other. As independence is not a requirement for LMM, this strengthens the choice of LMM as the analysis method. Furthermore, contrary to LRA, LMM considers respondent-specific baseline values (intercepts) in the analysis by including a random effect in the model, meaning that it can take into account, e.g., respondent’s different stress levels, before there are any explanatory variables for them (Diez Roux, 2002). This is important because assuming that each individual would have the same baseline value could lead to biased results. Besides, LMM can be used to analyze not only random effects but also fixed effects in the same model, which LRA is incapable of (Ibrahim et al., 2010). Finally, LMM can handle respondents with missing data, whereas other methods, such as Repeated measures ANOVA, would exclude them (Schafer & Yucel, 2012). LMM holds assumptions of a random sample, sample size, data linearity, intraclass correlation (ICC), normality of residuals, homoscedasticity, and absence of multicollinearity. To demonstrate statistical significance, a cutoff of p-value < 0.05 was used. Before the analysis, we calculated composite scores for ASQ dimensions representing internal and external attention based on the corresponding items reported in Van Calster et al. (2018). Composite scores were calculated for BFI dimensions according to the author’s manual to build factors representing the five personality traits (Rammstedt & John, 2007). Furthermore, we created a new variable representing the sequential numbering of the day of data collection. The ordinal variable of data collection day (Day_of_collection) was created to enter it into the model as a covariate so that the effect of other variables (such as personality and metacognitive beliefs) on the dependent variable could be assessed independently of time. In addition to the data collection day variable, we created two variables, the first corresponding to the week number of the data collection day (Week_Number) and the second to the weekly average of the physiological stress values (Weekly_Stress_Level). Linear mixed models were created by first adding all fixed factors and covariates to the model and then examining the SPSS output to exclude the least statistically significant predictors one at a time. At the same time, the 25 improvement of the model fit was evaluated. All statistical analyses were conducted using IBM SPSS Statistics 28.0.1.1. 32 value(195,756), the present model could predict the variation in perceived stress scores better than the null model. TABLE 4 The multilevel linear mixed effects model of perceived stress with weekly averages. 4.6 Residuals and homoscedasticity of Model 2 To ensure that our data meets the assumptions of LMM, we will discuss the distribution of the residuals and homoscedasticity of the model represented in Table 4. The distribution of the residuals is illustrated in Figure 6. By investigating the figure, we observe that the residuals are concentrated evenly near the zero. Additionally, there are exceptionally small values on the right side of the histogram, which indicates that the distribution is slightly skewed to the right. The residuals are mainly concentrated near zero, and no significant outliers would violate the normality assumption. Fixed Effects Estimate Std. Est.* Std. Error Sig. 95% CI lower 95% CI upper Intercept 75,326 -1,288E-15 6,260 <,001 62,590 88,061 MCQ_total -,635 -,546 ,099 <,001 -,836 -,434 BFI_Extraversion -2,716 -,653 ,415 <,001 -3,559 -1.873 BFI_Conscientiousness -1,095 -,309 ,334 ,002 -1,775 -,415 Random Effects σ² σ² Intercept 0 0 2,020 <,001 Residual 8,206 ,219 0 Intraclass Correlation (ICC) 0 0 Model Fit (-2RLL) 182,164 57,419 Note. * Standardized Estimate 33 FIGURE 6 Distribution of the Residuals for Model 2. The normal Q-Q plot of the residuals is represented in Figure 7. Examination of the figure shows that the residuals visually fall on a straight line in the center, but the values at the beginning and end of the line deviate somewhat from the straight line. Based on these findings, we can estimate that with the lowest and highest values, the model's ability to predict physiological stress scores might be insufficient. The outliers at the line's end further confirm that the distribution is slightly skewed to the right. Despite the slight deviations with high and low values, the normality assumption is not violated with average values. 34 FIGURE 7 Normality Q-Q Plot of the Residuals for Model 2. Based on the examination of the homoscedasticity of the residuals shown in Figure 8, we observe that the residuals are not evenly distributed around zero. There is noticeable clustering at specific predicted value levels, particularly at the highest predicted values (around 24), where residual variance appears very small and consistent. This pattern suggests a potential issue with heteroscedasticity, as the spread of residuals is not uniform across the entire range of predicted values. Specifically, the variance of residuals is much smaller for the highest predicted values. These observations indicate potential problems with the homoscedasticity assumption. FIGURE 8 Examination of the Homoscedasticity for Model 2. 35 4.7 Personality, attentional styles, and metacognitive beliefs as interaction terms in Model 2 We conducted a multilevel linear mixed-effects model to determine whether the degree of personality, metacognitive beliefs, or attentional styles potentially affects the relationship between perceived stress scores (PSS-10_Stress) and physiological stress scores (Weekly_Stress_Level). As we noted with the previous model (See Table 4), only conscientiousness (p = .002), metacognitive beliefs (p = <.001), and extraversion statistically significantly (p = <.001) predicted perceived stress scores. Therefore, we rejected our third hypothesis regarding the moderating effect of attentional styles on the relationship between physiological and perceived stress. We tested hypotheses 2 and 4 concerning the potential moderating effect of personality and metacognitive beliefs by adding conscientiousness, extraversion, and metacognitive beliefs as interaction variables to the model. The weekly averaged physiological stress scores (Weekly_Stress_Level) were added to the model as a main effect. Neither the interaction variables nor physiological stress statistically significantly predicted perceived stress values in the model. Thus, hypotheses 2 and 4 were not supported in the study. 36 The current study used the linear mixed model to investigate the relationship between perceived and physiological stress scores and the potential moderating effects of attentional styles, metacognitive beliefs, and the Big Five personality traits. We created two models to test our hypotheses. The first predicted weekly values of perceived stress with daily values of physiological stress. In model 2, weekly averaged physiological stress values predicted weekly perceived stress scores. This chapter discusses the results, starting with the first model. The first model examined the predictability of perceived stress levels based on physiological stress levels, and the results showed that physiological stress scores did not predict perceived stress values. This result is contrary to the previous research. Föhr et al. (2015) found a positive correlation between the HRV-derived stress index calculated with the Firstbeat Analytics method and perceived stress measured with the PSS-14. Besides, Can et al. (2020) found that physiological measurements based on wrist-measured HRV predicted long-term perceived stress levels measured with the Perceived Stress Scale (PSS-14). Similarly, Martinez et al. (2022) reported that the triangular index and low- and highfrequency components of HRV were significant predictors of perceived stress. These discrepancies may be due to differences in sample size, the type of physiological measures used, the timing and frequency of data collection, or the specific aspects of stress being assessed. Although Can et al. (2020) also used a PPG sensor as a physiological measure in their study, the HRV-based stress measure might differ from Garmin's use of HRV to calculate stress values. While Martinez et al. (2022) used Garmin devices to measure HRV, they did not use the stress value calculated by the device to measure stress, so the physiological measurements might not be aligned with our study. In addition, Can et al. (2020) and Föhr et al. (2015) measured perceived stress with a 14-item scale of PSS, while in our study, it was measured weekly with a 10-item version of PSS. Thus, perceived stress measured in the previous studies may have been measured more comprehensively than in this study, which may reduce its predictability based on physiological stress. 5 DISCUSSION 37 Besides, Martinez et al. (2022) used ecological momentary assessment as a measure of stress in their study, which differed from the validated stress measure we used. In contrast to the previous findings, Fazeli et al. (2021) reported that there is growing evidence that subjective stress experience and physiological stress are disconnected. This disconnection may explain the divergence in our results compared to the studies discussed. It may reflect differences in how individuals perceive and report their stress levels compared to their physiological stress responses. However, Garmin's stress measure has yet to be validated as a physiological stress measure. Therefore, it is essential to acknowledge that this numerical stress value does not necessarily represent an absolute measure of a person's stress level. Along with these weaknesses, our sample only consisted of five people, in which case the amount of statistical power suffers compared to the larger sample size, which weakens the comparability of the results with previous research. Although we did not find a statistically significant relationship between physiological and perceived stress, we observed statistically significant associations in model 1 as exploratory findings. Specifically, agreeableness, conscientiousness, and neuroticism emerged as predictors of perceived stress scores, while metacognitive beliefs were close to statistical significance. Of the predictors, the estimate of conscientiousness was negative, indicating that as conscientiousness decreases, the perceived stress increases. This result aligns with the conclusion drawn in the meta-analysis by Luo et al. (2022), stating that conscientiousness was negatively related to perceived stress. In support of this, researchers (Ebstrup et al., 2011; Johnsen, 2013) have similarly found that conscientiousness correlates negatively with stress measured by the Perceived Stress Scale, which is consistent with our findings. This result may be explained by the mechanism that individuals with high conscientiousness scores may have developed strong coping mechanisms to manage the stress that arises from their conscientiousness, i.e., their tendency to complete tasks thoroughly. Since they have effective ways to cope with stress, they may not perceive it as intensely as those with weaker coping strategies. In past studies (Hamid et al., 2015; Penley & Tomaka, 2002), conscientiousness has been linked to problem-fo- cused coping strategies such as support-seeking and problem engagement. This offers a potential explanation for the negative relationship between conscientiousness and perceived stress. However, the dynamics might differ when we consider agreeableness and neuroticism, which were positively correlated with perceived stress. Our finding that agreeableness correlates positively with perceived stress contrasts with previous research (Ebstrup et al., 2011; Luo et al., 2022), which links agreeableness to lower perceived stress. In our data, however, more agreeable individuals appear to experience higher stress levels. It could be that agreeable persons perceive stress during conflicts. Because of that, they comply with others and experience stress because they do not act according to their own will, in which case their needs will not necessarily be satisfied. On the other hand, 38 since agreeable individuals tend to go along with others rather than impose their will (Costa et al., 1991), it can be theorized that agreeableness may be linked to lower stress perception, as it reduces the likelihood of conflict or confrontation. However, the mechanism between agreeableness and perceived stress would need more investigation to reveal the potential association and its nature. Along with agreeableness, the relationship between neuroticism and perceived stress was positive, suggesting that increasing scores on the personality trait would lead to higher perceived stress. The positive association we found between neuroticism and perceived stress aligns with prior research (Bolger & Schilling, 1991; Bolger & Zuckerman, 1995; Ebstrup et al., 2011; Jiang et al., 2017; Xin et al., 2017; Banjongrewadee et al., 2020; Luo et al., 2022), further supporting the established link between these variables. Given that neuroticism is associated with a tendency to experience negative emotions like anxiety, anger, and depression, it is unsurprising that individuals with higher levels of this trait also report elevated stress levels. This finding highlights how neuroticism can shape stress experiences, offering insight into individual differences in stress vulnerability. Nonetheless, the primary aim of this study was to examine whether the statistically significant predictors identified in the first model moderate the relationship between physiological and perceived stress. The multilevel model analysis revealed that the interaction terms of physiological stress with personality, metacognitive beliefs, and attentional styles did not statistically significantly explain perceived stress. We hypothesized that the Big Five personality traits would moderate the correlation between perceived and physiological stress since Luo et al. (2022) suggested that stress perceptions differ from physiological responses to stress in correlation strength and direction in individuals with specific personality traits. We hypothesized that attentional styles may moderate this relationship, as individuals who engage in internal, bottom-up attention are more prone to distraction by their thoughts, leading to rumination and greater sensitivity to stress perception (Van Calster et al., 2018). Additionally, researchers have found that rumination, which means focusing on current distress, positively correlates with stress (Nolen-Hoeksema, 1991; Willis & Burnett, 2016). Finally, we expected that metacognitive beliefs would moderate the relationship, as increased awareness of one’s thinking and cognition may heighten stress perception, particularly when individuals feel their knowledge or skills are insufficient to perform at a certain level. Metacognitive beliefs were reported to correlate positively with perceived stress in past studies (Ramos-Cejudo & Salguero, 2017; Sariçam, 2015; Spada et al., 2008). In this study, no variables were significant moderators in the statistical association examined. The insignificant moderating effects may indicate that the relationship between physiological and perceived stress does not vary meaningfully across levels of personality, metacognitive beliefs, or attentional styles. This suggests that these factors may influence perceived and physiological stress independently rather than altering the strength or direction of their association. The 39 result could imply that personality, cognitive, or attentional styles primarily affect stress perception or physiological reactivity in ways that do not directly shift the correlation between these stress indicators. Alternatively, the moderation effects might be more context-specific, emerging only under certain stress conditions or within particular settings where the potential moderators exert more substantial influence. Future studies could investigate whether these moderating effects become more pronounced in specific stress situations, where personality, attentional styles, or metacognitive beliefs more acutely shape stress appraisal and physiological response. Finally, it is essential to consider that physiological and perceived stress may capture distinct components of the stress experience. Physiological measures may reflect acute arousal responses, while perceived stress is influenced by cognitive appraisal and emotional response to stressors over time. This difference might limit how personality or cognitive styles moderate their relationship, as each stress measure reflects distinct underlying processes. Because we did not find significant moderating effects of the three variables when examining how weekly physiological stress predicted perceived stress, we re-examined these interactions in a second model using weekly averaged physiological stress scores to predict perceived stress. In this second model, we used only weekly values for predictions rather than daily values to predict weekly values as in the first model. This method was supposed to align values from the same time points, enhancing result comparability. The results of the second model indicated that weekly physiological stress scores did not predict perceived stress. Therefore, the result concerning the first hypothesis did not change when weekly values were used for prediction. However, we observed statistically significant connections in model 2 as exploratory findings. According to the results of model 2, conscientiousness, metacognitive beliefs, and extraversion statistically significantly predicted perceived stress scores. The exploratory findings differ from the results of model 1, so the metacognitive beliefs close to statistical significance became statistically significant in model 2. In addition, agreeableness and neuroticism fell below the statistical significance threshold, with a negative correlation of extraversion becoming statistically significant along with the previously significant conscientiousness. Of the statistically significant predictors in Model 1, conscientiousness was the most significant predictor, and agreeableness and neuroticism were the least significant. Thus, it is logical that in the week-level model, with fewer observations and thus less noise, strong correlations remained, and associations bordering on statistical significance became significant or non-significant. Hence, two novel significant correlations emerged from model 2 for discussion. The result that metacognitive beliefs are negatively associated with perceived stress is inconsistent with previous literature. Perceived stress levels have positively correlated to metacognitive beliefs in previous studies (Ramos- Cejudo & Salguero, 2017; Sariçam, 2015; Spada et al., 2008). According to our results, higher scores on metacognitive beliefs, i.e., an increased focus on one's 40 thinking and cognition, predict a lower perceived stress level. By reviewing the items of the MCQ-30, we notice that the scale measures both cognitive self-con- sciousness (e.g., "5. I am aware of the way my mind works when I am thinking through a problem" and "16. I am constantly aware of my thinking") and the extent of worrying (e.g., "4. I could make myself sick with worrying" and "11. I cannot ignore my worrying thoughts") (Wells & Cartwright-Hatton, 2004, p. 7). A high level of cognitive self-consciousness, such as understanding how one's mind works in problem-solving, could be linked to lower perceived stress. However, Spada et al. (2008) theorized that their contrasting finding of a positive association might stem from individuals who control their thoughts but not their worries, as they may be more likely to adopt maladaptive coping strategies—such as repetitive thinking patterns—that could increase their experience of stress. In summary, as we noticed that the scale has an emphasis on the statements of worrying tendencies, which are linked to higher stress experience (Kowalczyk et al., 2023), our contradicting result may be due to the limitations of this study, which will be discussed later. Finally, we found that extraversion is negatively correlated with perceived stress. Thus, individuals with a high extraversion trait experienced less stress. Similarly, a negative association has been found in previous studies (Ebstrup et al., 2011; Fernández-Mendoza et al., 2010; Kiekens et al., 2015). Ebstrup et al. (2011) suggested that the negative correlation between extraversion and perceived stress is likely because extroverts consider difficult events challenging rather than threatening. Therefore, extroverts may, as outgoing persons, see the situation as a challenge that is an honor to overcome and use emotional and problem-focused coping methods to talk to others and ask for help. Although this study did not investigate the coping methods used by the participants, we might assume that extroverts may have used these methods to limit their stress experience. Since conscientiousness, metacognitive beliefs, and extraversion were significant predictors of perceived stress in model 2, we tested if they had a moderation effect on the relationship under investigation. As in model 1, none of the variables statistically significantly moderated the association investigated. Thus, we rejected our hypotheses 2, 3, and 4. To summarize our rejected hypotheses, they were listed as follows: H1: A positive correlation exists between physiological and perceived stress. H2: There is a moderating effect of personality on the relationship between physiological and perceived stress. H3: Attentional styles have a moderating effect on the relationship between physiological and perceived stress. 41 H4: There is a moderating effect of metacognitive beliefs on the relationship between physiological and perceived stress. Next, we will discuss the limitations of this study. 5.1 Limitations The study has limitations related to the design, measurement methods, sample, and results, which will be critically discussed in this section. Although the association between physiological and perceived stress has been investigated in previous studies, there is no scientific consensus about the relationship between these factors. Several reasons can be listed for this discrepancy, which has potentially influenced the results of this study as well. Metsämuuronen (2001) has emphasized the importance of evaluating the internal and external validity of the study, namely intended construct measurability and generalizability of the results. This study used the PSS-10 scale developed by Cohen et al. (1983) to measure perceived stress. Several studies have confirmed its validity (Kechter et al., 2019; Lee, 2012; Liu et al., 2020; Xiao et al., 2023), and the scale is widely used for measuring perceived stress. Additionally, the Perceived Stress Scale was used in studies (Can et al., 2020; Föhr et al., 2015) to investigate stress perception in the same context as our study. Thus, several studies support the observation of the internal validity of the measure, and there is no reason to doubt that the measure would not be valid for measuring perceived stress. However, the optimal approach would have been to measure perceived stress daily, as physiological stress was measured, to improve the comparability and accuracy of the measurements. Nevertheless, this may have negatively affected participants' motivation to respond, as it would have required a greater time investment, potentially limiting the number of participants or the response rate. However, evaluating the validity of our measure of physiological stress is not as straightforward as that of perceived stress. We used Garmin smartwatches to measure physiological stress, which relies on Firstbeat's analysis method in the calculation. The method's validity in measuring physiological stress is debatable since studies have not validated Garmin’s Firstbeat analysis as a stress measure. Although the principles of the analysis method are explained in its original patent document (Kettunen & Saalasti, 2005), we cannot conclude precisely how the analysis works on each device, depending on its age. We also used different Garmin smartwatches in our research, so we cannot ensure their consistent measurement. Still, Garmin smartwatches’ HR measurement, from which the stress estimate is derived, has been validated in studies (Merrigan et al., 2022). Nevertheless, the literature suggests that the accuracy of HR measurements in smartwatches depends on the intensity level of the physical task performed while worn (Düking et al., 2020; Hajj-Boutros et al., 2021). Therefore, the 48 Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84(4), 822–848. https://doi.org/10.1037/0022-3514.84.4.822 Can, Y. S., Chalabianloo, N., Ekiz, D., Fernandez-Alvarez, J., Repetto, C., Riva, G., Iles- Smith, H., & Ersoy, C. (2020). Real-Life Stress Level Monitoring Using Smart Bands in the Light of Contextual Information. IEEE Sensors Journal, 20(15), 8721– 8730. https://doi.org/10.1109/jsen.2020.2984644 Cao, R., Rahmani, A. M., & Lindsay, K. L. (2022). Prenatal stress assessment using heart rate variability and salivary cortisol: A machine learning-based approach. PLOS ONE, 17(9), e0274298. https://doi.org/10.1371/journal.pone.0274298 Carciofo, R., Yang, J., Song, N., Du, F., & Zhang, K. (2016). Psychometric Evaluation of Chinese-Language 44-Item and 10-Item Big Five Personality Inventories, Including Correlations with Chronotype, Mindfulness and Mind Wandering. PLOS ONE, 11(2), e0149963. https://doi.org/10.1371/journal.pone.0149963 Chalmers, T., Hickey, B. A., Newton, P., Lin, C.-T., Sibbritt, D., McLachlan, C. S., Clifton-Bligh, R., Morley, J., & Lal, S. (2021). Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables. Sensors, 22(1), 151. https://doi.org/10.3390/s22010151 Charter, R. A. (1999). Sample Size Requirements for Precise Estimates of Reliability, Generalizability, and Validity Coefficients. Journal of Clinical and Experimental Neuropsychology, 21(4), 559–566. https://doi.org/10.1076/jcen.21.4.559.889 49 Chun, M. M., Golomb, J. D., & Turk-Browne, N. B. (2011). A Taxonomy of External and Internal Attention. Annual Review of Psychology, 62(1), 73–101. https://doi.org/10.1146/annurev.psych.093008.100427 Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A Global Measure of Perceived Stress. Journal of Health and Social Behavior, 24(4), 385–396. https://doi.org/10.2307/2136404 Cohen, S., & Williamson, G. M. (1988). Perceived stress in a probability sample of the United States. In S. Oskamp (Ed.), The Social Psychology of Health (pp. 31–67). SAGE. Cook, S. A., Salmon, P., Dunn, G., & Fisher, P. (2014). Measuring Metacognition in Cancer: Validation of the Metacognitions Questionnaire 30 (MCQ-30). PLoS ONE, 9(9), e107302. https://doi.org/10.1371/journal.pone.0107302 Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98–104. https://doi.org/10.1037/0021- 9010.78.1.98 Costa, P. T., McCrae, R. R., & Dye, D. A. (1991). Facet Scales for Agreeableness and Conscientiousness: A Revision of the NEO Personality Inventory. Personality and Individual Differences, 12(9), 887–898. https://doi.org/10.1016/0191- 8869(91)90177-d Deng, Z., Guo, L., Chen, X., & Wu, W. (2023). Smart Wearable Systems for Health Monitoring. Sensors, 23(5), 2479. https://doi.org/10.3390/s23052479 50 Derryberry, D., & Reed, M. A. (2002). Anxiety-related attentional biases and their regulation by attentional control. Journal of Abnormal Psychology, 111(2), 225–236. https://doi.org/10.1037/0021-843x.111.2.225 Dhingra, L. S., Aminorroaya, A., Oikonomou, E. K., Nargesi, A. A., Wilson, F. P., Krumholz, H. M., & Khera, R. (2023). Use of Wearable Devices in Individuals With or at Risk for Cardiovascular Disease in the US, 2019 to 2020. JAMA Network Open, 6(6). https://doi.org/10.1001/jamanetworkopen.2023.16634 Diez Roux, A. V. (2002). A glossary for multilevel analysis. Journal of Epidemiology & Community Health, 56(8), 588–594. https://doi.org/10.1136/jech.56.8.588 Düking, P., Giessing, L., Frenkel, M. O., Koehler, K., Holmberg, H.-C., & Sperlich, B. (2020). Wrist-Worn Wearables for Monitoring Heart Rate and Energy Expenditure While Sitting or Performing Light-to-Vigorous Physical Activity: Validation Study. JMIR MHealth and UHealth, 8(5), e16716. https://doi.org/10.2196/16716 Ebstrup, J. F., Eplov, L. F., Pisinger, C., & Jørgensen, T. (2011). Association between the Five Factor personality traits and perceived stress: is the effect mediated by general self-efficacy? Anxiety, Stress & Coping, 24(4), 407–419. https://doi.org/10.1080/10615806.2010.540012 Egeth, H. E., & Yantis, S. (1997). Visual Attention: Control, Representation, and Time Course. Annual Review of Psychology, 48(1), 269–297. https://doi.org/10.1146/annurev.psych.48.1.269 Epel, E. S., Crosswell, A. D., Mayer, S. E., Prather, A. A., Slavich, G. M., Puterman, E., & Mendes, W. B. (2018). More than a feeling: A unified view of stress 51 measurement for population science. Frontiers in Neuroendocrinology, 49(1), 146– 169. https://doi.org/10.1016/j.yfrne.2018.03.001 Fajkowska, M., & Derryberry, D. (2010). Psychometric properties of Attentional Control Scale: The preliminary study on a Polish sample. Polish Psychological Bulletin, 41(1), 1–7. https://doi.org/10.2478/s10059-010-0001-7 Fan, L., Yuan, D., Gao, C., Xiong, K., Geng, F., & Zhang, L. (2023). Validity and Reliability of the Metacognitions Questionnaire-30 (MCQ-30) among Chinese Adolescents. Child Psychiatry & Human Development. https://doi.org/10.1007/s10578-023-01625-7 Fazeli, S., Levine, L., Cascalheira, C. J., Baharan Mirzasoleiman, Zadeh, B., Peris, T. S., & Majid Sarrafzadeh. (2022). Passive Monitoring of Physiological Precursors of Stress Leveraging Smartwatch Data. 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm55620.2022.9995354 Fernández-Mendoza, J., Vela-Bueno, A., Vgontzas, A. N., Ramos-Platón, M. J., Olavarrieta-Bernardino, S., Bixler, E. O., & De la Cruz-Troca, J. J. (2010). Cognitive- Emotional Hyperarousal as a Premorbid Characteristic of Individuals Vulnerable to Insomnia. Psychosomatic Medicine, 72(4), 397–403. https://doi.org/10.1097/psy.0b013e3181d75319 Firstbeat Technologies Ltd. (2014, September 16). Stress and Recovery Analysis Method Based on 24-hour Heart Rate Variability. https://assets.firstbeat.com/firstbeat/uploads/2015/11/Stress-and-recovery_white-pa- per_20145.pdf 52 Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitivedevelopmental inquiry. American Psychologist, 34(10), 906–911. https://doi.org/10.1037//0003-066x.34.10.906 Föhr, T., Tolvanen, A., Myllymäki, T., Järvelä-Reijonen, E., Rantala, S., Korpela, R., Peuhkuri, K., Kolehmainen, M., Puttonen, S., Lappalainen, R., Rusko, H., & Kujala, U. M. (2015). Subjective stress, objective heart rate variability-based stress, and recovery on workdays among overweight and psychologically distressed individuals: a cross-sectional study. Journal of Occupational Medicine and Toxicology, 10(1). https://doi.org/10.1186/s12995-015-0081-6 Garson, D. (2020). The Null Model. In Multilevel Modeling: Applications in STATA, IBM SPSS, SAS, R, & HLM (pp. 57–100). SAGE Publishing. https://doi.org/10.4135/9781544319315.n3 Greene, S., Thapliyal, H., & Caban-Holt, A. (2016). A Survey of Affective Computing for Stress Detection: Evaluating technologies in stress detection for better health. IEEE Consumer Electronics Magazine, 5(4), 44–56. https://doi.org/10.1109/mce.2016.2590178 Hajj-Boutros, G., Landry-Duval, M.-A., Comtois, A. S., Gouspillou, G., & Karelis, A. D. (2021). Wrist-Worn Devices for the Measurement of Heart Rate and Energy Expenditure: A Validation Study for the Apple Watch 6, Polar Vantage V and Fitbit Sense. European Journal of Sport Science, 23(2), 1–36. https://doi.org/10.1080/17461391.2021.2023656 Hamid, A., Roohafza, H. R., Keshteli , A. H., Masood, M., Awat Feizi, A., & Peyman , A. (2015). The association of personality traits and coping styles according to 53 stress level. Journal of Research in Medical Sciences, 20(4), 353–353. https://doi.org/10.4103/1735-1995.158255 Head, A., Lo, K., Kang, D., Fok, R., Skjonsberg, S., Weld, D. S., & Hearst, M. A. (2021). Augmenting Scientific Papers with Just-in-Time, Position-Sensitive Definitions of Terms and Symbols. Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3411764.3445648 Hehlmann, M. I., Schwartz, B. S., Lutz, T., Martín, J., Rubel, J. A., & Lutz, W. (2021). The Use of Digitally Assessed Stress Levels to Model Change Processes in CBT - A Feasibility Study on Seven Case Examples. Front Psychiatry, 12(12). https://doi.org/10.3389/fpsyt.2021.613085 Hepburn, S.-J., Carroll, A., & McCuaig, L. (2021). The Relationship between Mindful Attention Awareness, Perceived Stress and Subjective Wellbeing. International Journal of Environmental Research and Public Health, 18(23), 12290. https://doi.org/10.3390/ijerph182312290 Hickey, B. A., Chalmers, T., Newton, P., Lin, C.-T., Sibbritt, D., McLachlan, C. S., Clifton-Bligh, R., Morley, J., & Lal, S. (2021). Smart Devices and Wearable Technologies to Detect and Monitor Mental Health Conditions and Stress: A Systematic Review. Sensors, 21(10), 3461. https://doi.org/10.3390/s21103461 Huang, C.-J., Chan, H.-L., Chang, Y.-J., Chen, S.-M., & Hsu, M.-J. (2021). Validity of the Polar V800 Monitor for Assessing Heart Rate Variability in Elderly Adults under Mental Stress and Dual Task Conditions. International Journal of Environmental Research and Public Health, 18(3), 869. https://doi.org/10.3390/ijerph18030869 54 Ibrahim, J. G., Zhu, H., Garcia, R. I., & Guo, R. (2010). Fixed and Random Effects Selection in Mixed Effects Models. Biometrics, 67(2), 495–503. https://doi.org/10.1111/j.1541-0420.2010.01463.x Jiang, J. M., Seng, E. K., Zimmerman, M. E., Sliwinski, M., Kim, M., & Lipton, R. B. (2017). Evaluation of the Reliability, Validity, and Predictive Validity of the Subscales of the Perceived Stress Scale in Older Adults. Journal of Alzheimer’s Disease : JAD, 59(3), 987–996. https://doi.org/10.3233/JAD-170289 John, O. P., Robins, R. W., & Pervin, L. A. (2008). Handbook of Personality: Theory and Research (3rd ed., p. 128). Guilford Press. Johns, M. W. (1991). A New Method for Measuring Daytime Sleepiness: The Epworth Sleepiness Scale. Sleep, 14(6), 540–545. https://doi.org/10.1093/sleep/14.6.540 Johnsen, S. Å. K. (2013). Exploring the use of nature for emotion regulation: Associations with personality, perceived stress, and restorative outcomes. Nordic Psychology, 65(4), 306–321. https://doi.org/10.1080/19012276.2013.851445 Katsuki, F., & Constantinidis, C. (2014). Bottom-up and top-down attention: Different processes and overlapping neural systems. The Neuroscientist, 20(5), 509–521. https://doi.org/10.1177/1073858413514136 Kechter, A., Black, D. S., Riggs, N. R., Warren, C. M., Ritt-Olson, A., Chou, C.-P., & Pentz, M. A. (2019). Factors in the Perceived Stress Scale Differentially Associate with Mindfulness Disposition and Executive Function among Early Adolescents. Journal of Child and Family Studies, 28(3), 814–821. https://doi.org/10.1007/s10826-018-01313-4 55 Kettunen, J., & Saalasti, S. (2005). Procedure for detection of stress by segmentation and analyzing a heartbeat signal. https://patentimages.storage.googleapis.com/c0/da/86/0013898102e400/US20050256414A1.pdf Kiekens, G., Bruffaerts, R., Nock, M. K., Van de Ven, M., Witteman, C., Mortier, P., Demyttenaere, K., & Claes, L. (2015). Non-Suicidal Self-Injury Among Dutch and Belgian Adolescents: Personality, Stress and Coping. European Psychiatry, 30(6), 743–749. https://doi.org/10.1016/j.eurpsy.2015.06.007 Kiyonaga, A., & Egner, T. (2012). Working memory as internal attention: Toward an integrative account of internal and external selection processes. Psychonomic Bulletin & Review, 20(2), 228–242. https://doi.org/10.3758/s13423-012-0359-y Kowalczyk, M., Kornacka, M., Kostrzewa, Z., & Izabela Krejtz. (2023). Differences in anxiety, worry, and perceived stress among naturally cycling women and oral contraceptives users: a cross-sectional study investigating the role of contraceptive types. Archives of Women’s Mental Health, 27(2). https://doi.org/10.1007/s00737-023-01405-1 Kunst, A. (2023, November 29). Wearable device ownership 2021. Statista. https://www.statista.com/forecasts/1101101/wearable-devices-ownership-in- selected-countries Leary, M. R., & Hoyle, R. H. (2013). Handbook of Individual Differences in Social Behavior (pp. 129–146). Guilford Publications. Lee, E.-H. (2012). Review of the psychometric evidence of the Perceived Stress Scale. Asian Nursing Research, 6(4), 121–127. https://doi.org/10.1016/j.anr.2012.08.004 56 Lee, I., Park, N., Lee, H., Hwang, C., Kim, J. H., & Park, S. (2021). Systematic Review on Human Skin-Compatible Wearable Photoplethysmography Sensors. Applied Sciences, 11(5), 2313. https://doi.org/10.3390/app11052313 Li, K., Cristiano, Moctezuma-Ramirez, A., Abdelmotagaly Elgalad, & Perin, E. C. (2023). Heart Rate Variability Measurement through a Smart Wearable Device: Another Breakthrough for Personal Health Monitoring? International Journal of Environmental Research and Public Health, 20(24), 7146–7146. https://doi.org/10.3390/ijerph20247146 Lim, Y. I., & Pratt, J. (2022). The interaction of internal and external attention. Attention, Perception, & Psychophysics, 85(2). https://doi.org/10.3758/s13414-022-02577-1 Lin, L. I-Kuei. (1989). A Concordance Correlation Coefficient to Evaluate Reproducibility. Biometrics, 45(1), 255. https://doi.org/10.2307/2532051 Liu, X., Zhao, Y., Li, J., Dai, J., Wang, X., & Wang, S. (2020). Factor Structure of the 10- Item Perceived Stress Scale and Measurement Invariance Across Genders Among Chinese Adolescents. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.00537 Lovik, A., Nassiri, V., Verbeke, G., Molenberghs, G., & Sodermans, A. K. (2017). Psychometric properties and comparison of different techniques for factor analysis on the Big Five Inventory from a Flemish sample. Journal of Psychological and Educational Research, 25(2), 122–129. https://doi.org/10.1016/j.paid.2017.05.048 Lu, Y., & Zhou, Z. (2024). Study on the Relationship Between Technology, Data and Body Isomorphism—The Example of Smartwatches. Telematics and Informatics Reports, 13, 100118–100118. https://doi.org/10.1016/j.teler.2024.100118 57 Luo, J., Zhang, B., Cao, M., & Roberts, B. W. (2022). The Stressful Personality: A Meta- Analytical Review of the Relation Between Personality and Stress. Personality and Social Psychology Review, 27(2), 108886832211040. https://doi.org/10.1177/10888683221104002 Maher, C., Ryan, J., Ambrosi, C., & Edney, S. (2017). Users’ experiences of wearable activity trackers: a cross-sectional study. BMC Public Health, 17(1). https://doi.org/10.1186/s12889-017-4888-1 Martinez, G. J., Grover, T., Mattingly, S. M., Mark, G., D’Mello, S., Aledavood, T., Akbar, F., Robles-Granda, P., & Striegel, A. (2022). Alignment Between Heart Rate Variability From Fitness Trackers and Perceived Stress: Perspectives From a Large-Scale In Situ Longitudinal Study of Information Workers. JMIR Human Factors, 9(3), e33754. https://doi.org/10.2196/33754 McCrae, R. R., & John, O. P. (1992). An introduction to the five-factor model and its applications. Journal of Personality, 60(2), 175–215. https://doi.org/10.1111/j.1467- 6494.1992.tb00970.x Merrigan, J. J., J. Hannah Stovall, Stone, J. D., Stephenson, M., Finomore, V., & Hagen, J. A. (2022). Validation of Garmin and Polar Devices for Continuous Heart Rate Monitoring During Common Training Movements in Tactical Populations. Measurement in Physical Education and Exercise Science, 27(3), 234–247. https://doi.org/10.1080/1091367x.2022.2161820 Metsämuuronen, J. (2001). Delfi-tutkimuksen reliabiliteetti. In Sosiaali- ja terveysalan tulevaisuutta etsimässä. International Methelp Ky. 64 5. Olen täyttänyt 18 vuotta. -Kyllä -Ei 6. Minulla ei ole diagnosoituna jotakin seuraavista terveydellisistä tai mielenterveydellisistä häiriöistä: alkoholismi, ahdistushäiriö, tai jokin vakava unihäiriö, kuten unettomuus. -On diagnosoitu -Ei ole 7. Minulla on jokin seuraavista lääkityksistä: säännöllinen kipulääkitys, rauhoittava lääkitys, tai verenpainelääkitys. -Kyllä -Ei 8. Ole hyvä ja syötä sähköpostiosoitteesi, jotta voimme generoida sinulle linkin tutkimukseen osallistumista varten. Sähköposti tulee syöttää molempiin alla oleviin kenttiin. - - 65 APPENDIX 2 BASELINE QUESTIONNAIRE 8. Mikä on ikäsi? (vuosina) - 9. Mikä on sukupuolesi? -Nainen -Mies -Muu -En halua sanoa 9. Mikä on pituutesi? (senttimetreinä) -Alle 150cm -150-154cm -155-159cm -160-164cm -165-169cm -170-174cm -175-179cm -180-184cm -185-189cm -190-194cm -195-199cm -200-204cm -Yli 205cm 10. Mikä on painosi? (kilogrammoina) -Alle 50kg -50-54kg -55-59kg 66 -60-64kg -65-69kg -70-74kg -75-79kg -80-84kg -85-89kg -90-94kg -95-99kg -100-104kg -105-109kg -110-114kg -115-119kg -Yli 120kg 11. Mikä on syntymävuotesi? - 12. Mikä on syntymäkuukautesi? - Tammikuu - Helmikuu - Maaliskuu - Huhtikuu - Toukokuu - Kesäkuu 67 - Heinäkuu - Elokuu - Syyskuu - Lokakuu - Marraskuu - Joulukuu 13. Mikä on siviilisäätysi? - Naimaton - Naimisissa - Avoliitossa - Asumuserossa tai eronnut - Leski 14. Missä kaupungissa asut? - 15. Mikä on korkein koulutuksesi? - Peruskoulu/oppikoulu/kansakoulu - Ammattikurssi, oppisopimus tms. - Ammattikoulu - Keskiasteen opisto tai lukio - Alempi korkeakoulututkinto - Ylempi korkeakoulututkinto - Tohtorin tutkinto tai vastaava 16. Mikä on työtilanteesi? - Työssä kokopäiväisesti - Työssä osapäiväisesti 68 - Työttömänä - Lomautettuna - Eläkkeellä - Opiskelija - Omaa kotitaloutta hoitava (kotiäiti tai -isä) - En halua kertoa 17. Onko sinulla jokin liikuntaa rajoittava seikka (fyysinen vamma tai muu)? Vastaa kyllä/ei ja kerro tarkemmin. Jos et halua kertoa, kirjoita laatikoon "En halua kertoa". -Kyllä -Ei 18. Arvioi, kuinka paljon käytät urheiluun vuodessa? (Sis. välineet ja aktiviteetit) Arvioi summa myös, jos jokin muu osapuoli kustantaa kulusi! - 0-100 euroa - 100-500 euroa - 500-1000 euroa - 1000-2000 euroa - 2000-5000 euroa - yli 5000 euroa 19. Mihin luokkaan luokittelet itsesi urheilullisen profiilisi suhteen? - Olen kilpaurheilija - Harrastan liikuntaa tavoitteellisesti ja säännöllisesti - Harrastan liikuntaa säännöllisesti - En harrasta liikuntaa säännöllisesti 20. Mikä on pääasiallinen urheilulajisi TAI yleisin kuntoilu- tai liikuntamuotosi? - 69 21. Korkein taso, jossa olet kilpaillut? 1. Harrastetaso 2. Alueellinen taso 3. Kansallinen taso 4. Kansainvälinen taso 22. Kuinka monta vuotta olet harrastanut urheilua? - 23. Kuinka monta tuntia viikossa harjoittelet lajisi parissa? Jos et harrasta tiettyä lajia, valitse vaihtoehto 1. 24. Minulla ei ole erityistä lajia, jonka parissa harjoittelen. 2. 1-5 tuntia 3. 5-10 tuntia 4. 10-15 tuntia 5. Yli 15 tuntia 25. Kuinka paljon hyötyliikuntaa harrastat viikossa? Tähän kuuluvat esimerkiksi työmatkaliikunta, kotityöt, pihatyöt ja portaiden kävely. 1. Alle 1 tunti 2. 1-5 tuntia 3. 5-10 tuntia 4. 10-15 tuntia 5. Yli 15 tuntia 70 26. Kuinka paljon kuntoilua tai muuta liikuntaa harrastat viikossa? Tähän kuuluvat esimerkiksi lenkkeily, kuntosali, erilaiset pallopelit tai muut fyysisesti kuormittavat harrastukset. Laske mukaan myös mahdollinen lajin parissa käytetty aika. 1. Alle 1 tunti 2. 1-5 tuntia 3. 5-10 tuntia 4. 10-15 tuntia 5. Yli 15 tuntia 27. Kuinka tärkeäksi koet urheilukellon käytön osana harjoitteluasi? -En lainkaan tärkeänä -Jonkin verran tärkeänä -Erittäin tärkeänä 28. Kuinka usein tarkkailet kellostasi unta/aktiivisuuttasi/palautumistasi? -En tarkkaile lainkaan -1-2 kertaa viikossa -Useita kertoja viikossa, kerran päivässä -Useita kertoja viikossa, monta kertaa päivässä 29. Valitse värikartasta väri, joka miellyttää sinua eniten. Siirrä väriympyrässä sijaitsevaa pistettä muuttaaksesi väriä, ja väriympyrän alla olevaa pistettä muuttaaksesi värin tummuutta. Valitsemasi väri näkyy väriympyrän viereisessä laatikossa. 71 30. Valitse värikartasta väri, jonka koet vähiten miellyttäväksi. Siirrä väriympyrässä sijaitsevaa pistettä muuttaaksesi väriä, ja väriympyrän alla olevaa pistettä muuttaaksesi värin tummuutta. Valitsemasi väri näkyy väriympyrän viereisessä laatikossa. 30. Palkitsemme aktiivisimmat vastaajat leffalipuilla. Ilmoitamme palkkion tiedot tulorekisteriin, jolloin tarvitsemme henkilötietosi. Tiedot kysytään myöhemmin, ne salataan eikä niitä yhdistetä kerättyyn dataan. Haluatko tulla palkituksi aktiivisuudesta? -Kyllä -Ei 31. Syötä tähän kenttään matkapuhelinnumerosi. Matkapuhelinnumeroasi hyödynnetään sähköpostin ohella välittämään linkit tutkimuksen kyselyihin sekä itsearviointiin. Numero tulee syöttää molempiin alla oleviin kenttiin. - - 72 APPENDIX 3 ATTENTIONAL STYLE QUESTIONNAIRE Answer the following statements. Choose the most suitable option. Rated on a 6-point Likert scale: 1 = Totally disagree, 2 = Disagree, 3 = Somewhat disagree, 4 = Somewhat agree, 5 = Agree, 6 = Totally agree 1. I have trouble concentrating when there is movement in the room I am in. 2. In general, I stay in control of my thoughts and do not let myself get distracted by interfering thoughts. 3. I am easily drawn to new stimuli (for example, voices of people passing by, a sound in the house, ...) that are not relevant to a task I am doing. 4. I can be so absorbed by a line of thoughts that I become more or less unaware of my surroundings. 5. When I am doing a task, I am often so focused I do not notice my surroundings. 6. I do not have difficulties to work while listening to music. 7. It is hard for me to stay on one activity for a whole hour. 8. During an activity, unrelated mental images and thoughts come to my mind. 9. I often put hold to an activity because I suddenly think about another one I have to start or continue. 10. I generally stay focused on a single task until it is finished. 11. I can easily ignore my surroundings. 12. Sometimes I interrupt an activity to check an unrelated detail. 13. When I am working on my computer, I often go on the internet to visit websites that are unrelated to my work. 14. I can easily concentrate on a task, even when there is movement in the room I am in. 15. I can spend several minutes on a question and try to dissect it. 73 16. I have trouble thinking when there are noises, even if these noises are not intense. 17. I am often the first one to notice something has changed in a room.