Investigating the neurophysiological markers of ability beliefs

(1)

Investigating the neurophysiological markers of ability beliefs

36 EC

21 August 2017 – 23 February 2018

Name of Student : Farah Aulia Rahman Student ID : 11392584

Supervisor : dr. Tieme W.P. Janssen

Co-assessor : dhr. dr. Wery van den Wildenberg UvA representative : dhr. dr. Wery van den Wildenberg

Research Institute : Vrije Universiteit Amsterdam

MSc Brain and Cognitive Sciences, University of Amsterdam Cognitive Neuroscience track

(2)

Abstract

In educational environments, the implicit belief that a student holds about their own abilities contributes to their academic level of achievement. Individuals with an incremental mindset see intelligence as something malleable and can improve with effort, whereas individuals with an entity mindset see intelligence as rigid and fixed. Many research has investigated behavioural differences between the mindsets, however the physiological mechanisms underlying this effect has not been sufficiently explored. To address this, we investigated the baseline measurement and reactivity of effort-related physiological measures: skin conductance, pre-ejection period (PEP), respiratory sinus arrhythmia (RSA), and alpha band power, along with the behavioural patterns during a Math Effort Task. Findings confirmed that individuals with an entity mindset behaved in a way that displayed lack of confidence in their own abilities whereas those with an incremental mindset were persistent in improving their ability. As for the physiological measures, only PEP reactivity was found to be significantly different between the incremental group and entity group.

(3)

1. Introduction

In educational environments, academic abilities is not the only determinant of a student’s level of achievement. The implicit belief that a student holds about the nature of their own intelligence has proven to also have predictive power over their achievements and learning motivation (Blackwell, Trzesniewski, & Dweck, 2007). Generally, people hold on to one of two beliefs regarding their intelligence and abilities. Dweck’s theory of intelligence (TOI) (Dweck & Leggett, 1988) categorise those who view intelligence as a rigid entity which cannot be changed as having an entity belief (also popularly known as a fixed mindset) while those who believe that intelligence is malleable and improves with effort are categorised as having an incremental belief (also known as a growth mindset). Students with an incremental mindset have often been reported to obtain higher grades and academically perform better than their counterparts (Blackwell et al., 2007; Good, Aronson, & Inzlicht, 2003).

Considering the impact that mindset has on the outcome of education, researchers have begun to investigate potential interventions to encourage a more incremental mindset. Reading scientific articles of how the mind is either immutable or malleable was found to have the capability to induce different mindsets (Bergen, 1991 cited in Schroder, Moran, Donnellan, & Moser, 2014). Students who were told that the brain can be trained analogous to the muscle have displayed better grades than those who didn’t (Blackwell et al., 2007). However, the change of mindset in these studies were measured solely through changes in questionnaire results. Measures of mindset change has started to shift towards physiological measures in a study measuring ERPs to characterize the two mindsets (Schroder et al., 2014). Having a deeper understanding of the underlying behaviours and processes of having different implicit beliefs could therefore lead researchers to a reliable measure of the effectivity of a mindset intervention.

Research has suggested that the effect of mindset towards academic achievement is mediated by differences in self-regulatory processes which contribute to academic motivation and learning (Schunk & Ertmer, 2000). Burnette states that goal setting and goal operating are among the core processes of self-regulation which are linked to implicit theories (Burnette, O’Boyle, VanEpps, Pollack, & Finkel, 2013). Goal setting involves establishing the desired final result that a person aims to achieve. Incremental theorists generally pursue learning goals, therefore striving to improve their abilities and learn to master the task at hand. Entity theorists pursue performance

goals, aiming to showcase their ability through successes and ultimately perform better than others.

The strategies implemented to achieve these goals are referred to as the goal operating process. Incremental theorists adopt mastery-oriented strategies such as investing more effort in order to improve their abilities while entity theorists tend to adopt helpless-oriented strategies such as self-handicapping strategies which protects them from the appearance of being incapable when mistakes are made (Diener & Dweck, 1980).

Characteristics between mindsets were not only found in behaviour but also reflected in brain responses. Mangels, Butterfield, Lamb, Good, and Dweck (2006) hypothesized that because entity theorists are more performance-oriented, negative feedback for mistakes that they have made will affect them more greatly than incremental theorists who see mistakes as learning opportunities. The research found that entity theorists showed enhanced attentional ERP responses for performance-relevant feedback (correct or incorrect) while incremental theorists displayed higher attentional ERP responses for learning-relevant feedback (the correct answer of an incorrect response). Having a different mindset was also found to affect attention allocation between stimuli- and response-processing (Schroder et al., 2014), error-positivity ERP (Pe) amplitudes in school children (Schroder et al., 2017), and level of awareness towards mistakes (Moser et al., 2011).

A numerous amount of research has been aimed at investigating the behavioural characteristics of holding a certain implicit belief (Blackwell et al., 2007; Hong, Chiu, Dweck, Lin, & Wan, 1999; Kammrath & Peetz, 2012) and an increasing number of research have identified neurocognitive signature responses (Mangels et al., 2006; Moser et al., 2011; Schroder et al., 2017, 2014), however

(4)

little is known about the physiological processes underlying the different mindsets. To the knowledge of the researcher, investigation into the autonomous differences between the two mindsets however have not yet been explored although evidence linking characteristic mindset behaviours to physiological responses exist. Individuals with an entity mindset have been suggested to be more vulnerable to stress and anxiety than individuals with an incremental mindset when faced with challenges (Howell, 2016) and stress has often been correlated to various autonomous responses such as skin conductance (Jacobs et al., 1994; Lazarus, Speisman, & Mordkoff, 1963) and heart rate variability (Sun et al., 2012). Individuals with an incremental mindset have been observed to invest more effort into a task as compared to those with an entity mindset (Dweck, 1986; Hong et al., 1999) and research has suggested that effort levels are reflected in sympathetic and parasympathetic activity (Harper, Eddington, & Silvia, 2016) as well as brain activity (Brouwer, Hogervorst, Holewijn, & van Erp, 2014).

Both stress and effort have correlated physiological responses through regulations by the autonomous nervous system. Stress reactions are regulated by the sympathetic nervous system responsible for fight-or-flight responses. Regular findings have determined that increased skin conductance levels (SCL) are highly correlated to high levels of stress (Jacobs et al., 1994; Wijsman, Grundlehner, Liu, Hermens, & Penders, 2011). Similarly, the sympathetic nervous system also affects cardiovascular responses which indicate the investment of effort, one of which being the pre-ejection period (PEP) (Gendolla, Wright, & Richter, 2012). Faster PEP responses have been said to correlate with higher effort engagement (Kelsey, 2011). Although effort is more commonly detected through sympathetic measurements, recent research related to effort has started investigations into parasympathetic-regulated cardiovascular responses as well (Harper et al., 2016; Silvia, Eddington, Beaty, Nusbaum, & Kwapil, 2013). Heart rate variability has previously been demonstrated to indicate effort (Althaus, Mulder, Mulder, van Roon, & Minderaa, 1998) and respiratory sinus arrhythmia (RSA) is a component of heart rate variability which increases with higher task engagement (Harper et al., 2016). Since effort is a mental process, it is not surprising that it is also reflected in measures of the central nervous system. Multiple research has mentioned alpha suppression as a reliable index of effort (Brouwer et al., 2014; Gevins & Smith, 2003; Keil, Mussweiler, & Epstude, 2006). With evidence of connections between mindset and effort and established links between effort and physiological responses, there is possibility for finding direct relationships between mindset and physiological responses.

The current exploratory study aims to investigate the behavioural patterns of having different mindsets and their potential stress- and effort-related physiological markers during a mathematical task where difficulty levels are self-adjusted at each session. Based on previous research, the following hypothesis can be made regarding the behaviour of the different mindsets. It is predicted that incremental theorists will begin with a difficulty level similar to their assessed level of math ability and gradually choose more difficult levels towards the last sessions of the task whereas entity theorists will consistently opt for levels lower than their math ability level throughout the multiple sessions of the task. Due to the limited literature available on physiological indicators of mindsets, only a few hypotheses can be made about the physiological measures. Entity theorists are predicted to display higher SCL compared to incremental theorists as an indicator of experiencing more stress and anxiety. It is also predicted that they will show shorter PEP, smaller increases in RSA, and higher alpha band power as indicators of low effort. Based on a review suggesting that RSA and RSA reactivity may reflect different aspects (Beauchaine, 2015), physiological reactivity for SCL, PEP, and RSA will also be explored.

2. Methods 2.1. Participants

The data for this study was obtained from participants of a larger study (BrainBeliefs, ERC project ID: 716736) investigating the mechanisms of how beliefs and goals shape resilience to setbacks and achievements at school. The sample consisted of N=26 (8 males, 18 females)

(5)

undergraduate students enrolled in psychology or pedagogy courses at Vrije Universiteit Amsterdam with a mean age of M=20.41 (SD=2.22) years. All participants reported having normal or corrected to normal vision. Each participant filled out and signed an informed consent form prior to the experiment and were compensated financially or given research credits for their participation.

2.2. Questionnaire

Each participant was sent an overview of the research procedure and a website link to the online questionnaires (Qualtrics) which had to be filled in at least 48 hours prior to the time of the experiment. The questionnaires were available in English and Dutch and was a combination of several different constructs aimed to measure a variety of qualities: Theory of Intelligence (TOI) (Castella & Byrne, 2015) questionnaire to measure ability beliefs; Behavioural Inhibition System and Behavioural Approach System Scales (BIS/BAS) (Franken, Muris, & Rassin, 2005) to measure regulation of aversive and appetitive motives; Positive and Negative Affect Schedule (PANAS) scales (Watson, Clark, & Tellegen, 1988) to measure mood factors; Neuroticism, Extraversion, and Openness Personality Inventory Revised (NEO-PI-R) (Hoekstra, Ormel, & Fruyt, 2007) to measure neuroticism and extraversion; Achievement Goal Questionnaire-Revised (AGQ-R) (Elliot & Murayama, 2008) to measure achievement goals; and Self- Efficacy (Midgley et al., 2005) questionnaire to measure self-perception of competence. For the purpose of the current study, only the TOI questionnaire results will be used for analysis. The TOI questionnaire consists of sixteen statements serving the purpose of indicating the ability belief that the participant holds. For each statement, participants were required to indicate their agreement on a 6-point Likert scale (1 = strongly disagree, 6 = strongly agree). The statements indicate a tendency towards either an incremental belief (e.g. I believe I have the ability to change my basic level of cleverness considerably over time) or an entity belief (e.g. I don't think I personally can do much to increase my cleverness). Assessment of helplessness (Crandall, Katkovsky, & Crandall, 1965) and positive and negative effort (Licht & Dweck, 1984) was also measured after the Math Effort Task (see below). See Appendix for the list of statements for all questionnaires.

2.3. Math Effort Task

The Math Effort Task (MET) (Engle-Friedman et al., 2003) was used to assess the mental effort through behaviour patterns. In this task, participants are required to do 50 mathematical problems divided over five rounds. Each mathematical equation is a combination of four arithmetic operations: summation, subtraction, multiplication, and/or division, and are available in five levels of difficulty, depending on the combination of these operations. The task begins with three practice trials and a basic ability session where the participants are required to solve 20 mathematical problems (four problems of each level of difficulty) in ascending order of difficulty level. The task then continues with five rounds of mathematical problems consisting of 10 problems each. At the beginning of each round, the participant must choose the level of difficulty that they prefer for the next 10 problems. The differences in difficulty are determined by the combination of arithmetic operations and number magnitudes. Level 1 includes summation and subtraction of three single-digit numbers; Level 2 includes summation and subtraction of two single- and one double-single-digit numbers; Level 3 includes summation and subtraction of one single- and two double-digit numbers; Level 4 includes one multiplication operation, one summation operation, and one subtraction operation with at least three two-digit numbers out of four numbers; Level 5 includes one multiplication operation, one division operation, and one summation or subtraction operation of four two-digit numbers. The order of operations rule, which explains that multiplication and division operations are to be calculated before summation and subtraction, is briefly explained before the basic ability session begins. Example equations of each difficulty is presented in Table 1. The maximum score (when all mathematical problems are answered correctly) that can be attained at each round is also determined by the chosen level of difficulty. The highest score possible is 10 which can only be obtained if the participant chooses Level 5 and answers all problems correctly. The highest score is 9 for Level 4; 8 for Level 3; 7 for Level 2; and 6 for Level 1.

(6)

For each mathematical problem, the participant has ten seconds to respond using a number pad. The answer to each problem ranges from 0 to 9, therefore the participant only needs to respond by pressing a single button corresponding to the answer that they believe is correct.

Table 1. Example problems of each difficulty level in the Math Effort Task

Level Example problem

Level 1 8 - 4 + 5 Level 2 7 - 13 + 12 Level 3 8 - 67 + 64 Level 4 (3 * 2) + 70 - 69 Level 5 38 * 25 / 50 - 12 2.4. Psychophysiological measurements

Electroencephalograph (EEG) data was collected continuously at a sampling rate of 512 Hz using the ActiveTwo Biosemi system and ActiView software at 128 electrodes across the scalp, with neither a low or high pass filter. Electrooculogram (EOG) electrodes were placed above and below the left eye to measure vertical eye movements and horizontally at the outer canthus of the left and right eye to measure horizontal eye movements. As reference points, two electrodes were placed at both mastoids.

Cardiovascular and skin conductance measurements were recorded using the VU Ambulatory Monitoring System (VU-AMS). Electrocardiogram (ECG) was measured using three electrodes placed at three locations on the abdomen: below the right collarbone, between the two lowest ribs on the right side, and approximately at the apex of the heart. Impedance cardiogram (ICG) was measured using two electrodes placed on the front of the abdomen (at the upper tip and lower tip of the sternum) and the back (on the spine, 3 cm higher than the electrode on the upper tip of the sternum and 3 cm lower than the electrode on the lower tip of the sternum). Skin conductance levels were measured by two electrodes placed on the palm of the left hand.

2.5. Procedure

The participants were asked to read and sign an informed consent form as well as fill in a short questionnaire on their caffeine intake (in coffee and/or tea) and alcohol prior to the appointment and their quantity and quality of sleep on the night prior to the day of the experiment. The EEG recording cap, ECG, ICG, and electrodermal electrodes were then attached to the participant and recorded physiological responses continuously throughout the entire experiment. The experiment started with two control measurements: one control session with eyes closed for a duration of 5 minutes and the next with eyes open while watching a documentary video of plants (Geffen & Williams, 2012) without audio for 10 minutes. The participants were then required to undergo the MET for approximately 15 minutes and afterwards fill in a questionnaire regarding their feeling of helplessness and positive and negative effort in regard to the mathematical task. This is then followed by a stop signal task (SST) lasting for a duration of approximately 40 minutes.

2.6. Data processing

In order to investigate the physiological differences between individuals with an incremental and entity mindset, the sample was first split into an incremental or entity mindset group. Questionnaire statements for the TOI questionnaire which lean towards an entity mindset were scored inversely so that a higher overall score would indicate towards an incremental mindset and lower scores would indicate towards an entity mindset. The sample is then categorized into two groups according to their mindset through a median split.

SCL is measured by the VU-AMS device and the average SCL over each session is automatically calculated by the VU-DAMS software. PEP is determined as the time interval between the Q-onset of the QRS complex on the ECG graph and the B-point on the dZ/dt graph (Krohova et al., 2017).

(7)

For each condition, a Large Scale Ensemble Average is computed across the duration of h session for the purpose of improving the signal quality. Both the Q-wave onset and B-point are detected automatically by the VU-DAMS software yet also visually inspected for necessary corrections and results in an averaged PEP value for each session. RSA is calculated using the peak-to-valley method (de Geus, Willemsen, Klaver, & van Doornen, 1995) where the shortest inter-beat interval (IBI) during inspiration is subtracted from the longest IBI during expiration. The VU-DAMS software automatically detects the shortest and longest IBIs for each inspiration-expiration phase using an automatic scoring algorithm. If either the shortest inspiratory or longest expiratory IBI is missing or a negative value is obtained, the RSA value for these breaths are determined as zero.

Processing of the EEG data was conducted using Brain Vision Analyzer 2 software. A band-pass filter of 0.1 to 30 Hz and a notch filter at 50 Hz were applied. EEG and EOG electrodes were re-referenced to the average of the signal obtained from the reference electrodes at the left and right mastoids. Removal of ocular artefacts was performed using independent component analysis (ICA). A whole-head approach was used to determine the location on the scalp where alpha band power activity is the strongest during MET.

2.7. Data analysis

Statistical analyses were performed with SPSS Version 25. Significance was assumed if p < 0.05. For behavioural measures, a 2 (mindset: entity vs. incremental) × 2 (task stage: average first two sessions – basic math ability vs. average last two sessions – basic math ability) two-way mixed ANOVA was conducted three times for analysis of the chosen difficulty level, percentage of correctness, and final scores. For each physiological measure (skin conductance, PEP, RSA, and EEG), analyses were conducted to compare whether any differences existed in baseline measurements (actual measurement values), task-related reactivity (the difference between MET average and baseline), or time-related reactivity (the average of the last 2 levels subtracted by the average of the first 2 levels). Baseline measurements were analysed through an independent t-test (entity vs. incremental) conducted separately for the control session and the MET session. Two-way mixed ANOVA is again used to analyse task-related reactivity, contrasting two dimensions: mindset (entity vs. incremental) and condition (control session vs. MET session), as well as to analyse time-related reactivity, also contrasting two dimensions: mindset (entity vs. incremental) and time point (average of first two sessions of MET vs. average of last two sessions of MET).

3. Results

Due to the limited number of participants, the sample was divided into an ‘entity’ group or ‘incremental’ group through a median split which resulted in 14 samples categorized into the ‘entity’ group with a mean score of 62.71 (SD = 7.38) and 12 samples categorized into the ‘incremental’ group with a mean score of 82.25 (SD = 9.8). Analyses were then conducted to compare the behavioural and physiological results between the two groups.

3.1. Behavioural measures

Chosen difficulty level

Independent t-test was used to examine the difference in the basic math ability of the two mindset groups. Incremental theorists were found to have no difference in math ability (M = 3.3) compared to entity theorists (M = 3.27), F(1,24) = 0.56, p = 0.9. Two-way mixed ANOVA was conducted on the influence of mindset (entity and mindset) on the chosen level of difficulty at two different time points (beginning and end of task). The main effect of mindset yielded an F ratio of F(1, 24) = 0.33, p = 0.57, indicating that no difference was found in the overall chosen level difficulty between entity theorists (M = 2.8) and incremental theorists (M = 3.00). The main effect of time point however yielded an F ratio of F(1, 24) = 7.10, p = 0.014, indicating a significant difference of chosen level difficulty between the beginning (M = 2.71) and end of the task (M = 3.09). A significant interaction effect was found, F(1, 24) = 4.32, p = 0.49, therefore post hoc tests were conducted to further investigate the interaction between mindset

(8)

and time point. On average, entity theorists chose significantly easier levels in the beginning of the math task (M = 2.46, SD = 0.25) in comparison to the end of the task (M = 3.14, SD = 0.26), t(13) = -2.85, p = 0.014, whereas for incremental theorists, there was no significant difference between the level of difficulty chosen at the beginning (M = 2.96, SD = 0.27) of the task and at the end of the task (M = 3.04, SD = 0.28), t(11) = -0.62, p = 0.55.

Percentage of correctness

No significant effect of mindset was found on the percentage of correct answers, F(1, 24) = 0.14, p = 0.72. There was also no significant influence of the time point of the task, F(1, 24) = 2.76, p = 0.30. No significant interaction effect was found, F(1, 24) = 2.76, p = 0.11.

Final score

No significant effect was found for either mindset, F(1, 24) = 0.02, p = 0.9, indicating that no difference was found for entity theorists (M = 56.34) or incremental theorists (M = 55.79). Analysis of time point also did not yield any significant effects, F(1, 24) = 0.77, p = 0.39, meaning that no difference was found in the comparison between average scores of the first two sessions of the task (M = 57.26) and the last two sessions of the task (M = 54.88). There was also no interaction effect between the two types of variables, F(1, 24) = 0.09, p = 0.77.

Figure 1. Behavioural trends of entity theorists (red line) and incremental theorists (blue line) in their choices of math difficulty level (top left), percentage of correct answers (top right), and average score (bottom)

3.2. Skin conductance level (SCL)

Baseline

Values of measurements during the control session and MET task were compared using separate independent t-tests. The graph (Figure 2) shows that during the control session, SCL appeared to be higher in incremental theorists (M = 8.48, SD = 5.24) than entity theorists (M = 5.4, SD = 4.3). The same pattern was found for the MET session where incremental theorists were found to have higher SCL (M = 11.36, SD = 5.53) than their counterparts (M = 8.34, SD = 5.54). However, this difference was found to be insignificant for both the control session average t(24) = -1.65, p = 0.11 and MET average t(24) = -1.39, p = 0.18.

(9)

Task-related reactivity

The two-way mixed ANOVA conducted resulted in a significantly higher SCL during the MET session (M = 9.85) than the control session (M = 6.94), F(1, 24) = 49.21, p < 0.01. This indicates that working on the MET may have caused arousal in the subjects compared to the control session, regardless of their mindset. However, no significant difference in SCL was found between entity theorists (M = 6.87) and incremental theorists (M = 9.92), F(1, 24) = 2.36, p = 0.14, and no interaction was found between the mindset and condition, F(1, 24) = 0.005, p = 0.943.

Time-related reactivity

Two-way mixed ANOVA showed that there was a significant effect of time point towards the SCL reactivity, F(1, 24) = 15.92, p = 0.001. Mindset, however, did not prove to have any effect on time-related reactivity of SCL, F(1, 24) = 0.002, p = 0.96, and no interaction effect was found, F(1, 24) = 0.86, p = 0.36.

Figure 2. Average skin conductance level during the control session and the separate sessions of the MET (left) and average skin conductance reactivity (MET session – baseline) of entity theorists (red line) and incremental

theorists (blue line)

3.3. Pre-ejection period (PEP)

Baseline

Independent t-tests of PEP measurements during the control session showed that no significant difference was found between entity theorists (M = 110.08, SD = 18.74) and incremental theorists (M = 114.29, SD = 13.31), t(24) = -0.65, p = 0.52. The average of PEP measurements of entity theorists during the MET session (M = 108.08, SD = 19.46) also did not differ significantly from incremental theorists during the same session (M = 111.06, SD = 14.49), t(24) = -0.44, p = 0.67.

Statistical analysis showed that in general there was a significantly faster PEP during the MET (M = 109.57) than during the control session (M = 112.18), F(1, 24) = 10.37, p = 0.004. Entity theorists however did not show any significant difference in PEP (M = 109.08) to incremental theorists (M = 112.67), F(1, 24) = 0.3, p = 0.59, and no interaction between mindset and condition was found, F(1, 24) = 0.57, p = 0.46. This indicates for both entity and incremental theorists, more effort was involved during the MET session compared to the control session.

No significant difference was found between PEP reactivity during the first two sessions of the MET (M = -3.27) and PEP reactivity during the last two sessions of the MET (M = -2.39), F(1, 24) = 2.85, p = 0.10. Having an entity mindset also did not produce significantly different PEP reactivity (M = -2.09) compared to having an incremental mindset (M = -3.57), F(1, 24) = 0.736, p = 0.4. A near significant interaction was found between mindset and time point of the

(10)

MET, F(1, 24) = 3.71, p = 0.07, therefore post hoc tests were conducted separately for entity theorists and incremental theorists. The results showed that for entity theorists, the PEP reactivity (PEP during the task – PEP during the control session) at the beginning of the task (M = -2.02, SD = 4.77) did not differ significantly to the PEP reactivity at the end of the task (M = -2.15, SD = 3.83), t(13) = 0.19, p = 0.85. However, a significant difference was found for incremental theorists, t(11) = -2.25, p = 0.046, where the PEP reactivity at the beginning of the task was significantly larger (M = -4.51, SD = 5.79) than the PEP reactivity at the end of the task (M = -2.63, SD = 3.76). This may indicate that incremental theorists gradually decrease the effort that they put into the task with time.

Figure 3. Average PEP during the control session and the separate sessions of the MET (left) and average PEP reactivity (MET session – baseline) of entity theorists (red line) and incremental theorists (blue line)

3.4. Respiratory sinus arrhythmia (RSA)

Baseline

The RSA data was found to violate the normality assumption, therefore the data was first logarithmically transformed. Log RSA during the control session did not seem to differ between entity theorists (M = 1.82, SD = 0.15) and incremental theorists (M = 1.83, SD = 0.2), t(24) = -0.14, p = 0.89). Similar results were found for the average MET session where log RSA was not significantly different between entity theorists (M = 1.83, SD = 0.1) and incremental theorists (M = 1.76, SD = 0.19), t(24) = 1.11, p = 0.28.

Insignificant differences were found between log of RSA during the control session (M = 1.829) and log of RSA during the MET session (M = 1.797), F(1, 24) = 1.50, p = 0.23. There was also no significant difference of RSA between entity theorists (M = 1.829) and incremental theorists (M = 1.799), F(1, 24) = 0.23, p = 0.64. No significant interaction was found between condition vs mindset, F(1, 24) = 1.99, p = 0.17.

The RSA reactivity in the first two sessions of MET (M = -4.65) did not differ significantly from the RSA reactivity in the last two sessions of MET (M = -6.30), F(1, 24) = 0.23, p = 0.64, indicating no effect of time-point on RSA reactivity. No effect was also found for mindset, where the RSA reactivity of entity theorists (M = 0.23) did not differ significantly from RSA reactivity of incremental theorists (M = -11.19), F(1, 24) = 1.89, p = 0.18. An interaction effect of mindset and time point was also not found, F(1, 24) = 1.06, p = 0.31.

(11)

Figure 4. Average RSA during the control session and the separate sessions of the MET (left) and average RSA reactivity (MET session – baseline) of entity theorists (red line) and incremental theorists (blue line)

3.5. Alpha band power

With the whole-head approach, the EEG signal for alpha waves was found to be strongest in the frontal area of the scalp (Figure 5), therefore statistical tests for alpha band power were conducted for data obtained from the frontal electrode (Fz).

Eyes open (EO) MET MET - EO

Figure 5. Mapping of alpha band power. Alpha band power was found to be relatively high in the frontal area during the eyes open (EO) session (left column) and significantly suppressed during the MET session (middle column). The

right column shows the magnitude of decreased alpha band power.

Baseline

Log of alpha band power during the control session for entity theorists (M = 0.28) was found to be insignificantly different from incremental theorists (M = 0.34), t(24), = -0.48, p = 0.64. Insignificant difference in alpha band power during the math task was also found between entity theorists (M = 1.83) and incremental theorists (M = 2.02), t(24) = -0.43, p = 0.67.

(12)

A significant effect of condition was found where the log of alpha band power was significantly higher during the control session (M = 0.31) compared to the MET session (M = 0.22), F(24) = 14.19, p = 0.001. However mindset seemed to have no significant effect on log of alpha band power, F(24) = 0.15, p = 0.7; incremental theorists (M = 0.29) and entity theorists (M = 0.25) had similar values. An interaction effect of mindset and condition was also not found, F(1, 24) = 0.38, p = 0.55.

Figure 6. Average alpha band power during the control session and the MET (left) and average alpha band power reactivity (MET – baseline) of entity theorists (red line) and incremental theorists (blue line)

4. Discussion

Considering the limited number of research on the physiological differences of having an incremental and entity mindset, the current study aimed to explore the behavioural patterns of implicit beliefs and the potentially corresponding effort-related physiological responses. Assessments were made to investigate potential differences in level choice patterns, performance, and final scores during a Math Effort Task. For SCL, PEP, RSA, and alpha band power, baseline measures were assessed for their potential as biomarkers and two types of reactivity was assessed: task-related reactivity, and time-related reactivity (excluding alpha band power).

The findings of the behavioural patterns did indeed display a difference in math difficulty level choice pattern between different mindsets although entity and incremental theorists were proven to have the same basic math ability. However, the obtained data contradicted the hypotheses made for both incremental and entity theorists. Incremental theorists appeared to start with a difficulty level that was approximately the same as their math level and stayed at the same consistent level of difficulty throughout the whole task. Although this was the predicted pattern for entity theorists, a stable choice could also reflect persistence in mastering a certain chosen level which reflects the mastery-oriented strategy (Dweck, 1986; Dweck & Leggett, 1988). It is possible that they will move on to a higher challenge once they feel they have mastered one level, which is difficult to achieve within only five sessions.

Entity theorists started the first session at a level that was significantly lower than their actual ability and gradually increased the chosen difficulty level to the same level as their assessed basic math ability. This provides evidence of entity theorists’ low self-confidence in their own abilities (Ommundsen, Haugen, & Lund, 2005). Entity theorists aim for performance goals (Burnette et al., 2013; Dweck, 1986), therefore it is understandable that out of the fear of performing poorly, entity theorists may have chosen levels lower than their math ability to ensure a higher probability of obtaining correct answers and consequently also better scores. With the current modified version of the MET, higher scores are also obtained by choosing higher levels, giving entity theorists the incentive to challenge themselves towards the end of the task when they have established self-confidence through a satisfactory performance.

(13)

No effect of time or mindset was found on the average final score, however this is believed to be a result of the score calculation theme applied in the current research. Scores obtained from getting few questions correct on a difficult level could also be obtained through more correct answers on an easier level. Therefore final scores were found to not be reflective of different strategies implemented by the two mindsets.

Incremental theorists were found to have an insignificant difference in SCL compared to entity theorists during both the control and MET sessions, therefore contradicting the hypothesis and SCL could not be considered as a biomarker. Regardless of mindset, heightened SCL measurements were observed during the MET session compared to the control session. As an indicator of mental stress and arousal (Jacobs et al., 1994), this result was not surprising. SCL has also been found to be indicative of task engagement (Pecchinenda, 1996), which of course would be higher during the MET than the control session. As the task progressed, skin conductance levels gradually decreased most likely as a result of individuals becoming more accustomed and familiar towards the task, causing less stress and arousal. Due to the insignificant difference of baseline PEP found between the incremental and entity group, PEP could not be said as a biomarker of mindset. Mindset was predicted to have an effect on reduction of PEP as an indicator of increased effort and this hypothesis was confirmed in the obtained results. Both the entity group and the incremental group displayed a reduction in PEP duration during the MET relative to the control session, showing that the task was difficult enough to trigger mental effort. Interestingly, the PEP values in the incremental group dropped significantly at the beginning of the task and gradually increased back towards its initial value towards the end of the task. Meanwhile, the entity group displayed a relatively mild reduction in PEP duration and stayed at a constant PEP value throughout the MET session. PEP may possibly be sensitive to the amount of effort invested in a task, therefore a higher reduction in PEP shows that incremental theorists invested more effort (Kelsey, 2011). Towards the end of the task, less effort was invested; not due to loss of interest or attention, but possibly as an indicator of mastery. In the behavioural data it was discovered that incremental theorists opted for the same difficulty level throughout the five sessions. The gradual lengthening of PEP duration therefore could mean that for the same repeated level of math difficulty, less and less effort was required. The findings of the current study may have uncovered evidence for the potential of PEP as a measure of effort intensity.

RSA did not seem to differ significantly between the incremental group and the entity group, failing it as a potential biomarker of mindset. However, a reduction in RSA was observed for both groups in during the MET session relative to the control session. RSA reduction from baseline measurements reflects heightened attention (Beauchaine, 2015), which could be considered a way of increasing the effort towards a certain task. The reduction was found to be larger for incremental theorists, which could reflect higher attention allocation as a strategy to finish the task at hand compared to entity theorists who are less driven due to a combination of low self-esteem and lack of effort (Dweck, 1986; Ommundsen et al., 2005).

Alpha band power has been established as a reliable measure of effort (Brouwer et al., 2014; Gevins & Smith, 2003), therefore findings of suppressed alpha power during the MET provided evidence that all participants exerted effort towards completing the task. The baseline alpha band power however was similar between the incremental and entity group, deeming it inadequate as a biomarker. Task-related reactivity also could not be considered an identifier of mindset due to the insignificant difference found between the two groups. Nevertheless, considering that time-relevant reactivity was not investigated, there is still a possibility for alpha suppression to be another measure of effort intensity.

The results of the current study however are restricted by certain limitations. First of all, implicit beliefs were regarded as two separate groups although in reality, fixed mindset and growth mindset are the two extremes of a spectrum. This practice could significantly affect the significance of the results, therefore reducing its statistical power. The study was also limited in the fact that only a small sample size was obtained, also adding doubt to the validity of the significance of the results. Suggestions for future studies would be to conduct regression correlation between mindset scores and physiological

(14)

measurements provided with a sufficient number of participants to verify the results obtained in this study.

To conclude, SCL, PEP, RSA, and alpha band power did not serve as biomarkers of incremental and entity mindsets, however differences in behavioural patterns and physiological reactivity were indeed found. Mindsets appeared to have an effect on choice pattern of difficulty level and PEP reactivity.

References

Althaus, M., Mulder, L. J. M., Mulder, G., van Roon, A. M., & Minderaa, R. B. (1998). Influence of respiratory activity on the cardiac response pattern to mental effort. Psychophysiology, 35, 420– 430.

Beauchaine, T. P. (2015). Respiratory sinus arrhythmia: A transdiagnostic biomarker of emotion dysregulation and psychopathology. Current Opinion in Psychology, 3, 43–47.

https://doi.org/10.1016/j.copsyc.2015.01.017

Blackwell, K. L., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit Theories of Intelligence Predict Achievement Across an AdolescentTransition: A Longitudinal Study and an Intervention in Child. Child Development, 78(1), 246–263. https://doi.org/10.1111/j.1467-8624.2007.00995.x Brouwer, A. M., Hogervorst, M. A., Holewijn, M., & van Erp, J. B. F. (2014). Evidence for effects of

task difficulty but not learning on neurophysiological variables associated with effort.

International Journal of Psychophysiology, 93(2), 242–252.

https://doi.org/10.1016/j.ijpsycho.2014.05.004

Burnette, J. L., O’Boyle, E. H., VanEpps, E. M., Pollack, J. M., & Finkel, E. J. (2013). Mind-sets matter: A meta-analytic review of implicit theories and self-regulation. Psychological Bulletin,

139(3), 655–701. https://doi.org/10.1037/a0029531

Castella, K. De, & Byrne, D. (2015). My intelligence may be more malleable than yours: the implicit theories self-form is a better predictor of achievement and motivation. The European Journal of

the Psychology of Education, 30(3), 1–28.

https://doi.org/http://dx.doi.org/10.1007/s10212-015-0244-y

Crandall, V. C., Katkovsky, W., & Crandall, V. J. (1965). Children’s Beliefs in Their Own Control of Reinforcements in Intellectual-Academic Achievement Situations. Child Development, 36(1), 91–109.

de Geus, E. J. C., Willemsen, G. H. M., Klaver, C. H. A. M., & van Doornen, L. J. P. (1995). Ambulatory measurement of respiratory sinus arrhythmia and respiration rate. Biological Psychology, 41(3), 205–227. https://doi.org/10.1016/0301-0511(95)05137-6

Diener, C. I., & Dweck, C. S. (1980). An analysis of learned helplessness: II. The processing of success.

Journal of Personality and Social Psychology, 39(5), 940–952.

https://doi.org/http://dx.doi.org/10.1037/0022-3514.39.5.940

Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist1, 41(10), 1040– 1048.

Dweck, C. S., & Leggett, E. L. (1988). A social cognitive approach to motivation and personality.

Psychological Review, 95(2), 256–273. https://doi.org/10.1037/0033-295X.95.2.256

Elliot, A. J., & Murayama, K. (2008). On the Measurement of Achievement Goals: Critique, Illustration, and Application. Journal of Educational Psychology, 100(3), 613–628. https://doi.org/10.1037/0022-0663.100.3.613

Engle-Friedman, M., Riela, S., Golan, R., Ventuneac, A. M., Davis, C. M., Jefferson, A. D., & Major, D. (2003). The effect of sleep loss on next day effort. Journal of Sleep Research, 12(2), 113–124. https://doi.org/10.1046/j.1365-2869.2003.00351.x

Franken, I. H. A., Muris, P., & Rassin, E. (2005). Psychometric properties of the Dutch BIS/BAS scales. Journal of Psychopathology and Behavioral Assessment, 27(1), 25–30.

(15)

https://doi.org/10.1007/s10862-005-3262-2

Gendolla, G. H. E., Wright, R. A., & Richter, M. (2012). Effort Intensity: Some Insights From the

Cardiovascular System. The Oxford Handbook of Human Motivation.

https://doi.org/10.1093/oxfordhb/9780195399820.013.0024

Gevins, A., & Smith, M. E. (2003). Neurophysiological measures of cognitive workload during human–computer interaction. Theoretical Issues in Ergonomics Science, 4(1–2), 113–131. https://doi.org/10.1080/14639220210159717

Good, C., Aronson, J., & Inzlicht, M. (2003). Improving adolescents’ standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental

Psychology, 24(6), 645–662. https://doi.org/10.1016/J.APPDEV.2003.09.002

Harper, K. L., Eddington, K. M., & Silvia, P. J. (2016). Perfectionism and effort-related cardiac activity: Do perfectionists try harder? PLoS ONE, 11(8), 1–11.

https://doi.org/10.1371/journal.pone.0160340

Hong, Y., Chiu, C., Dweck, C. S., Lin, D. M.-S., & Wan, W. (1999). Implicit theories, attributions, and coping: A meaning system approach. Journal of Personality and Social Psychology, 77(3), 588– 599. https://doi.org/10.1037/0022-3514.77.3.588

Howell, A. J. (2016). Implicit theories of personal and social attributes: Fundamental mindsets for a science of wellbeing. International Journal of Wellbeing. https://doi.org/10.5502/ijw.v6i3.529 Jacobs, S. C., Friedman, R., Parker, J. D., Tofler, G. H., Jimenez, A. H., Muller, J. E., … Stone, P. H.

(1994). Use of skin conductance changes during mental stress testing as an index of autonomic arousal in cardiovascular research. American Heart Journal, 128(6), 1170–1177.

https://doi.org/10.1016/0002-8703(94)90748-X

Kammrath, L. K., & Peetz, J. (2012). You promised you’d change: How incremental and entity theorists react to a romantic partner’s promised change attempts. Journal of Experimental

Social Psychology, 48(2), 570–574. https://doi.org/10.1016/J.JESP.2011.10.015

Keil, A., Mussweiler, T., & Epstude, K. (2006). Alpha-band activity reflects reduction of mental effort in a comparison task: A source space analysis. Brain Research, 1121(1), 117–127.

https://doi.org/10.1016/J.BRAINRES.2006.08.118

Krohova, J., Czippelova, B., Turianikova, Z., Lazarova, Z., Tonhajzerova, I., & Javorka, M. (2017). Preejection Period as a Sympathetic Activity Index: a Role of Confounding Factors. Physiol. Res,

66, 265–275. Retrieved from www.biomed.cas.cz/physiolres

Lazarus, R. S., Speisman, J. C., & Mordkoff, A. M. (1963). The Relationship Between Autonomic Indicators of Psychological Stress: Heart Rate and Skin Conductance. Psychosomatic Medicine,

25(1), 19–30. https://doi.org/10.1097/00006842-196301000-00004

Licht, B. G., & Dweck, C. S. (1984). Determinants of academic achievement: The interaction of

children’s achievement orientations with skill area. Developmental Psychology, 20(4), 628–636. https://doi.org/10.1037/0012-1649.20.4.628

Mangels, J. A., Butterfield, B., Lamb, J., Good, C., & Dweck, C. S. (2006). Why do beliefs about intelligence influence learning success? A social cognitive neuroscience model. Social Cognitive

and Affective Neuroscience, 1(2), 75–86. https://doi.org/10.1093/scan/nsl013

Midgley, C., Maehr, M. L., Hicks, L., Roeser, R., Urdan, T., Anderman, E. M., … Middleton, M. J. (2005). Patterns of Adaptive Learning Survey. What Do Children Need to Flourish

Conceptualizing and Conceptualizing and Measuring Indicators of Positive Development, 223–

235. https://doi.org/10.1007/0-387-23823-9

Moser, J. S., Schroder, H. S., Heeter, C., Moran, T. P., Lee, Y., Moser, J. S., … Lee, Y. (2011). Mind Your Errors : Evidence for a Neural Mechanism Linking Growth Mind-Set to. Psychological

Science, 22(12), 1484–1489. https://doi.org/10.1177/0956797611419520

Ommundsen, Y., Haugen, R., & Lund, T. (2005). Academic self-concept, implicit theories of ability, and self-regulation strategies. Scandinavian Journal of Educational Research, 49(5), 461–474. https://doi.org/10.1080/00313830500267838

(16)

Pecchinenda, A. (1996). The Affective Significance of Skin Conductance Activity During a Difficult Problem-solving Task. Cognition & Emotion, 10(5), 481–504.

https://doi.org/10.1080/026999396380123

Schroder, H. S., Fisher, M. E., Lin, Y., Lo, S. L., Danovitch, J. H., & Moser, J. S. (2017). Neural

evidence for enhanced attention to mistakes among school-aged children with a growth mindset.

Developmental Cognitive Neuroscience, 24, 42–50. https://doi.org/10.1016/j.dcn.2017.01.004

Schroder, H. S., Moran, T. P., Donnellan, M. B., & Moser, J. S. (2014). Mindset induction effects on cognitive control : A neurobehavioral investigation. Biological Psychology, 103, 27–37.

Schunk, D. H., & Ertmer, P. A. (2000). Regulation and Academic Learning. In Handbook of

Self-Regulation (pp. 631–649). https://doi.org/10.1016/B978-012109890-2/50048-2

Silvia, P. J., Eddington, K. M., Beaty, R. E., Nusbaum, E. C., & Kwapil, T. R. (2013). Gritty people try harder: Grit and effort-related cardiac autonomic activity during an active coping challenge.

International Journal of Psychophysiology, 88(2), 200–205.

https://doi.org/10.1016/j.ijpsycho.2013.04.007

Sun, F., Kuo, C., Cheng, H., Buthpitiya, S., Collins, P., & Griss, M. (2012). Activity-aware Mental Stress Detection Using Physiological Sensors. Mobile Computing, Applications, and Services, 76, 1–20. https://doi.org/10.1007/978-3-642-29336-8_16

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology,

54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063

Wijsman, J., Grundlehner, B., Liu, H., Hermens, H., & Penders, J. (2011). Towards mental stress detection using wearable physiological sensors. Conference Proceedings : ... Annual

International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, 2011, 1798–801.

(17)

Appendix

Caffeine-, alcohol-, and sleep-related questionnaire

English version

1. Today, did you have caffeinated drinks (such as coffee, cappuccino)? o Yes, number of drinks: ……… o No

2. Did you have tea today, or drinks having theine in it?

o Yes, number of drinks: ……… o No

3. Did you have alcoholic drinks last evening?

o Yes, number of drinks: ……… o No

4. On a normal day, how many caffeinated drinks do you have? Number of drinks: ………

5. On a normal day, how many drinks with theine do you have (such as tea)? Number of drinks: ………

6. How many hours of sleep did you have last night?

Number of drinks: ……… 7. Please rate the quality of sleep last night

o Very bad o Bad o Sufficient o Good o Very good

(18)

Post-MET questionnaires

Dear participant,

You just finished the arithmetic task. We would like to know how you reacted to the task. Below you see several statements that can be answered on a scale from 1 (strongly disagree) to 6 (strongly agree).

What did you think after you’ve made an error? Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree The task is confusing

o

I do not like the task

o

I’m not smart enough

o

I’m just not good in

arithmetic

o

What did you do or what happened after you’ve made an error? Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree I’ve putted more effort

into the task

o

I paid more attention towards the task

o

I hoped that it was finished soon

o

I’ve putted less effort

into the task

o

I lost my

(19)

Questionnaire TOI

The following statements are about you. Indicate the degree to which you disagree (on the left) or agree (on the right) with the following statements: Strongly disagree Disagree Somewhat disagree Somewhat

agree Agree Strongly agree

My aim is to completely master the material presented in this class

o

I don't think I personally can do much to increase how clever I am

o

It's much more important for

me to learn things in my classes than it is

to get the best grades

o

To be honest, I don't think I can

really change how good I am in statistics

o

My goal is to learn as much as possible

o

When something is hard, it just makes me want to work more on it, not less

o

I'm certain I can master the skills taught in class

this year

o

I can learn new things, but I don't have the ability to change my basic level of cleverness

o

(20)

If an assignment is hard, it means I'll probably learn a lot doing

it

o

I am striving to avoid performing worse than others

o

I am striving to avoid an incomplete understanding of the course material

o

With enough time and effort I

think I could significantly improve my statistic skills

o

I don't think I personally can do much to increase my statistical skills

o

If you're not good at a subject, working hard won't make

you good at it

o

My goal is to avoid learning less than what is

possible

o

My aim is to avoid doing worse than other students

o

Page Break

(21)

It doesn't matter how hard you work -

if you're not smart, you won't do well

o

How clever I am is something about me that I personally can't change very much

o

I believe I can always substantially improve on how clever I am

o

I believe I have the ability to change the basic

level of how clever I am considerably

over time

o

I'm certain I can figure out how to do the most difficult class work

o

I can do even the hardest work in this class if I try

o

Regardless of the current level

of how clever I am, I think I have the capacity to change it quite a bit

o

I am striving to do well compared to other students

o

Page Break

(22)

I can do almost all the work in

class if I don't give up

o

If you're not doing well at something, it's better to try something easier

o

My aim is to avoid learning less than I possibly could

o

Even if the work is hard, I can

learn it

o

If a subject is hard for me, it

means I probably won't be able to do really well at it

o

If I knew I wasn't going to do well at a task, I probably wouldn't do it even if I might learn a lot from

it

o

If you don't work hard and

put in a lot of effort, you probably won't do well

o

I believe I have the ability to change my statistical skills considerably over time

o

Page Break

(23)

My statistical skills are something about me that I personally can't change very much

o

I am striving to understand the content as thoroughly as possible

o

Regardless of my current skills in statistics, I think I have the

capacity to change it quite a bit

o

My aim is to perform well relative to other students

o

I can learn new things, but I don't have the

ability to my basic level of statistics

o

My goal is to avoid performing poorly compared to others

o

Although I hate to admit it, I sometimes would rather do well in a class than learn a lot

o

To be honest, I don't think I can

really change

how smart I am

o

(24)

I believe I can always substantially improve on my statistical skills

o

With enough time and effort I

think I could significantly improve how

clever I am

o

To tell the truth, when I work

hard at my schoolwork, it makes me feel like I'm not very

smart

o

My goal is to perform better than the other

students

o

The harder you work at something, the better you will

be at it

o

Page Break

If I had to choose between getting a good grade and being challenged in class, I would choose…

o

Good grade

o

Being challenged End of Block: TOI + AGQ