Running head: LEVELS OF IMPULSIVITY AND STRESS
Levels of Impulsivity and Stress on Accuracy Scores and Metacognition
Amanda Buck
St. Bonaventure University
Abstract
This
study assessed the difference between how individuals with two different types
of impulsivity react to stress (Pallier, Wilkinson, Danthiir, Kleitman,
Knezevic, Stankov, & Roberts, 2002); differences between their metacognitive abilities were also measured The
results indicated that functional impulsives had both higher accuracy
scores and more accu
Levels of Impulsivity and Stress on Accuracy Scores and Metacognition
Impulsivity is
“the tendency to deliberate less than most people of equal ability before
taking action” (Dickman, 1990). Impulsiveness is often seen as a negative
personality trait. As described by one researcher, impulsive individuals have a
small attention span that causes them to be counterproductive (Caplan &
Shechter, 1990). However, the consequences that follow impulsive acts are not
always negative. In fact, research shows that when given a simple task, an
individual who is high in impulsivity has a rapid response time and produces
little errors. Therefore, accuracy
scores are unaffected (Dickman, 1990).
Impulsivity has
been primarily assessed in terms of extraversion. Impulsivity is the trait of extraversion that
directly affects the most basic memory and perceptual processes and the way in
which they are carried out (Dickman, 1990).
When arousal or stress is increased, extroverts tend to show higher
levels of performance (Corr & Kumari, 1998). Extroverts also
have faster response times for motor behaviors, which places them at an
advantage under pressure (Doucet & Stelmack, 2000). Recently, research has focused on the effects
of stress on impulsivity and on the distinction between two different types of
impulsivity: functional and dysfunctional.
Dysfunctional
impulsivity has been widely studied and is defined as “a tendency to initiate
behavior before a situation has been analyzed adequately” (Caplan &
Shecter, 1990). Functional impulsivity
is defined as the tendency to act with relatively little forethought when such
a style is optimal (Dickman, 1990).
Previous research found that some traits related to functional
impulsivity are enthusiasm, activity, and adventurousness (Dickman, 1990). This finding is consistent with people’s
reports of benefiting from their impulsiveness (Dickman, 1990). These individuals are more productive and
take risks more often. It seems that when functional impulsives make a high
number of errors, it is offset by the amount of information they contribute to
the task (Dickman, 1990). Functionally
impulsive individuals work more quickly because they are intelligent (Dickman,
1990).
People who are dysfunctionally impulsive seem to work in the opposite way. They do not report beneficial outcomes attributed to their impulsivity, primarily because they are characterized by disorderliness and tend to ignore information necessary to make accurate decisions (Dickman, 1990). There is evidence that demonstrates that functional and dysfunctional impulsives react differently when under stress and have differences in cost during accuracy tasks (Dickman, 1990).
A stressful situation is a situation that is threatening or demanding in which necessary resources are unavailable to cope with the situation (Cohen, Kamarck & Mermelstein, 1983). The stressor that this study addresses is time pressure and its affect on accuracy of individual performance. Stress actually motivates performance (Corr & Kumari, 1998). Other effects that occur under time pressure are acceleration of individual thought processing when making decisions, greater attention to the information available before making a choice, and looking for simple strategies to encourage accurate decisions (Verplanken, 1993). These individuals also reported that it was more difficult to make a decision under stressful conditions (Verplanken, 1993).
Research also suggests that individuals, who are low in impulsiveness, are sensitive to stress-provoking stimuli, causing them to change their route of cognitive processing from automatic-processing to controlled-processing. In stressful situations, effort will increase, which causes cognitive performance to improve (Corr & Kumari, 1998). These findings imply that impulsive people are greatly affected by stress. In addition, dysfunctionally impulsive people tend to have difficulty processing information: They will have more of a deficit in accuracy when put under pressure. Surprisingly, stress can create positive mood states on certain occasions if the stressor is viewed as a challenge and not a threat (Folkman & Moskowitz, 2000). This would be characteristic of functional impulsives. Because these individuals are enthusiastic and highly productive, they are more accurate in making decisions under stressful situations. Because functional impulsives are characterized by being more productive, they will be less affected by stress than will dysfunctional impulsives.
Accuracy in
solving arithmetic tasks is measured in this study. Research proposed that stress causes
participants to narrow their focus of attention and have a difficult time
performing complex tasks (Zautra, Reich, Davis, Potter & Nicolson.,
2000). Research has illustrated that
functional impulsives are faster at processing information (Brunas-Wagstaff,
Bergquist and Wagstaff, 1994). In
stressful situations, functional impulsive individuals will benefit from their
impulsivity traits. Previous research reveals less of a deficit in accuracy scores
of participants (Brunas-Wagstaff et al. 1994). The underlying issue is that
there is a difference in the way individuals cope and use strategy when
completing stressful tasks (Becker & Schneider, 1976).
This study probes into the metacognitive abilities in functional and dysfunctional individuals. Research has suggested that personality plays a role in the accuracy of metacognition (Pallier, Wilkinson, Danthiir, Kleitman, Knezevic, Stankov, & Roberts, 2002). When subjects are faced with time pressure they are less confident about the decisions they make (Zautra et al., 2000). Time pressure affects confidence ratings. For this study, metacognition is defined as the predicted accuracy as determined by the participant in comparison with the actual score received on a mathematical task. Mathematical tasks deal with the cognitive ability of problem solving. Problem solving precipitates confidence biases because individuals tend to be over-confident about their knowledge (Pallier et al., 2002). In general, people are unaware of their own intellectual and social skills (Hacker, Bol, Horgan & Rakow, 2000). However, the more intelligent one is, the better he or she is able to predict metacognitive abilities accurately (Pallier et al., 2002). Functional impulsives are more confident and accurate than dysfunctional impulsives (Dickman, 1990). Confidence is an important factor in determining accuracy of self-assessment and it is therefore plausible to conclude that individuals who are functionally impulsive will be more accurate when making judgments about their performance.
In addition,
metacognition is developed in youth and improves over time. The more experience a person has in certain
fields, the better they are at predicting how well they did on a specific task
in a particular field. This is because individuals attribute past
outcomes to future results (Hacker et al., 2000).
After considering this information, it is hypothesized that when completing a mathematical task under stressful conditions, dysfunctionally impulsive individuals will complete the task with less accuracy than those who are functional impulsives. In addition, I am predicting an interaction such that functional impulsives will be more accurate in their percentage ratings for perceived scores (metacognition) and will be
under-confident, whereas dysfunctional impulsives will be less accurate in their percentage ratings for perceived scores (metacognition) and over-confident.
Method
Participants
Twenty undergraduates (10 functional impulsives and 10 dysfunctional impulsives) between the ages of 18-23 participated in this study. Each participant was placed in the stressful condition and in the non-stressful condition. Participants rated their performance according to the percentage that they thought they got correct on the math tasks after answering math questions in both conditions.
Materials
In order to assess the personality characteristic of impulsivity, the Eyesenck Personality Inventory Extraversion Scale (EPI-I Scale) was administered. Functional and dysfunctional impulsivity was measured by using the Dickman Impulsivity Inventory (DII) (Claes, Vertommen, & Braspenning, 2000). Stress was measured using the Perceived Stress Scale (PSS) (Cohen et al., 1983). This scale is designed to measure the degree of stress one experiences in his or her life in specific situations. A minor change in this scale was made for the purpose of this study. Instead of using the phrase “in the last month” it was modified to “during this task”. This scale is an accurate measure of life-event impact measures (Cohen, et al., 1983).
Two arithmetic
tasks were implemented in order to control for learning effects. Each consisted
of 15 problems, which were randomly. A short questionnaire was given to all
participants. The first question was,
“How many math courses have you taken at you University?” and “What percentage
of the items on the math tasks do you think you answered correctly?”
Procedure
The present study was completed over a 2-day period. On the first day, participants were given impulsivity tests. The Eyensenck Personality Inventory and the Dickman Impulsivity Inventory were used to determine impulsivity and the levels they exhibited. On the second day, the 10 participants who were functional impulsives were put into a stressful time pressure situation and told to complete a 15 question arithmetic task in 6 minutes. They were told to finish the questions as quickly as possible and to do their best. Every 2 minutes the participants were reminded of how much time they had to finish. In addition, the same participants were put into a situation that was not stressful in which the task was not timed. Participants were given unlimited time to finish the arithmetic task and told to take their time and do their best. The order of conditions was counterbalanced. After participating in each condition, participants completed the short questionnaire to determine their metacognitive abilities and to investigate whether there was a difference between metacognitive abilities under stressful versus non-stressful conditions. Dysfunctional impulsives also went through both of these conditions and were given the short questionnaire. The participants were given the PSS immediately after completing the questionnaire to ensure that participants in the time pressure conditions were experiencing stress. After the task was completed, the number of correct answers on the task was measured. This is the actual score.
Results
The results were analyzed using 3 separate 2 (functional
vs. dysfunctional impulsivity) x 2 (stress vs. no stress) analyses of variance
with actual scores, perceived scores, and the difference between the two,
serving as dependent variables in each.
The first analysis of variance measured actual scores. There was a main effect for type of impulsivity F (1, 18) = 38.926, p <.05. The results demonstrated that functionally impulsive individuals had higher actual scores than dysfunctionally impulsive individuals. There was also a main effect for stress level, or time pressure, F (1, 18) = 4.539, p < .05. The results illustrated that when stress was present, actual scores were lower and in a non-stressful situation, actual scores were higher. There was no interaction. This means that actual scores are not dependent upon type of impulsivity and stress level together. See Figure 1.
The second analysis of variance included impulsivity and stress with perceived scores serving as the dependent measure. There was a main effect for type of impulsivity F (1, 18) = 26.982, p < .05. This means that because functional impulsive individuals are characterized as being underconfident, they reported lower perceived scores than did the dysfunctional impulsives, who were overconfident. There was also a main effect for stress level F (1, 18) = 5.154, p < .05. This means that when in the stressful condition, individuals were less confident in predicting perceived scores than when in the non-stressful condition. Similar to the first analysis, there was no interaction, which means that perceived scores are not dependent upon type of impulsivity and stress level together.
In the third analysis of variance, individual perceived scores were subtracted from actual scores in order to determine metacognition, or how close they came to the real score when making predictions. These subtracted results were put into an analysis of variance. A main effect for type of impulsivity F (1, 18) = 16.012, p < .05, which shows that functional impulsives had higher metacognitive abilities. In other words the deviation between the perceived scores and actual scores was small. There was also a main effect for stress level F (1, 18) = 7.175), p < .05 were observed. Therefore, when individuals were placed under stress their metacognitive abilities decreased, showing a higher deviation from the actual score, than when in the non-stressful condition. There was no interaction for this analysis of variance. This means that metacognitive scores are not dependent upon type of impulsivity and stress level together.
The means for actual scores are presented in the appendix figure 1. The means for perceived scores are presented in the appendix figure 2, and the means for the difference between actual and perceived scores, or metacognition, are displayed in the appendix figure 3.
Discussion
Results of the main effects for both variables in the first
analysis of variance indicate that actual scores that participants gave
depended on their type of impulsivity and their level of stress. The second analysis of variance demonstrated
that the scores that participants predicted depended, once again, on their type
of impulsivity and level of stress. In
the third analysis, similar results were observed. Participants’ metacognitive
scores depended on their type of impulsivity and level of stress. In all analyses, however, no significant
interactions were observed. This
demonstrates that even though each variable had an effect on each one of the
dependent variables, they in no way depended upon each other in terms of
providing significant effects on the dependent variables.
In terms of metacognition, dysfunctional impulsives were less accurate in their percentage ratings. Under stressful conditions individuals were less confident in their judgments. Individuals who were more intelligent and had a history of doing well on tests, (functional impulsives) were under-confident in self-assessment and those who were less intelligent and did not do well on tests (dysfunctional impulsives) were overconfident in self-assessment (Palier et al., 2002).
Though the functional impulsives have high speed for information processing and showed some inaccurate answers, their method of processing was found to be “functional” under the specific conditions of the experiment. Functional impulsives had higher scores and higher accuracy when answering the questions (Dickman, 1990). Individuals who were dysfunctionally impulsive had lower accuracy scores.
The results showed that functional impulsives did in fact have higher accuracy scores than dysfunctional impulsives. There was also a difference in stress. When in a stressful situation accuracy scores were lower than when the participants were placed in a non-stressful situation. The same results were found using perceived scores and metacognitive scores. The hypotheses predicted main effects which were found. However, the hypothesis also predicted interactions, and this was not found.
There are some reasons the results did not completely coincide with the hypotheses made. For instance, the stressful condition may not have been stressful enough. The individuals may have been borderline dysfunctional or borderline functional. The extremes of this personality trait may have been necessary in order to result in a significant interaction. Some individuals may have been better in math than others and taken more math courses recently which could also pose a bias in the results.
In reference to metacognition, individuals generalize their cognitive abilities, causing them to predict consistent levels of confidence. Because of this pattern, confidence ratings are often inaccurate (Pallier et al., 2002). Very little research has been done to look at the effects of personality on metacognition. The present study implements procedures to investigate this growing area of research (Pallier et al., 2002). This may cause a bias in confidence ratings.
In future research it is suggested that participants be placed in a more stressful condition and extreme personalities are assessed. Metacognition could also be measured in different ways that would be more effective. For instance, one might look at how confident one is in his or her predictions, rather than only the perceived scores. A different task could also be used besides a math task to ensure that the situation was stressful. A different task could be used so that learning bias does not take place. If this is done properly it may be possible to find significant interactions.
References
Becker, P. & Schneider, E. (1976). Specific Reactions to Stress and Personality
Characteristics: Students Before an Examination. Aeitschrift fuer Experimentelle und Angewandte Psychologie, 23(1), 1-29.
Brunas-Wagstaff, J., Bergquist, A. & Wagstaff, G. F. (1994). Cognitive Correlates of
Functional and Dysfunctional Impulsivity in Adults. Personality and Individual Differences, 17(2), 289-292.
Caplan, B. & Shechter, J. (1990). Clinical Applications of the Matching Familiar Figures
Test: Impulsivity vs. Unilateral Neglect. Journal of Clinical Psychology, 46, 60-67.
Claes, L., Vertommen, H. & Braspenning, N. (2000). Factor Analysis and Description
of Psychometric Properties of Dickman Impulsivity Inventory in Dutch Speaking Adults. Personality and Individual Differences, 29(1), 27-35.
Cohen, S., Kamarck, T. & Mermelstein, R. (1983). A Global Measure of Perceived
Stress. Journal of Health and Social Behavior, 24, 385-396.
Corr, P. & Kumari, V. (1998). Impulsivity, Time of Day, and Stress: Effects on
Intelligence Test Performance. Journal of Research in Personality, 32, 1-12.
Dickman, S. (1990). Functional and Dysfunctional Impulsivity: Personality and
Cognitive Correlates. Journal of Personality and Social Psychology, 58, 95-102.
Doucet, C. & Stelmack, R. M. (2000). An Event-Related Potential Analysis of
Extraversion and Individual Differences in Cognitive Processing Speed and Response Execution. Journal of Personality and Social Psychology, 78, 956-964.
Folkman, S. & Moskowitz, J. T. (2000). Stress, Positive Emotion, and Coping.
Current Directions in Psychological Science, 9, 115-118.
Hacker, D. J., Bol, L., Horgan,
Performance in a Classroom Context. Journal of Educational Psychology, 92 (1),
160-170.
Pallier, G., Wilkinson, R., Danthir, V., Kleitman, S., Knezevic, G., Stankov, L. &
Roberts, R. D. (2002). The Role of Individual Differences in the Accuracy of Confidence Judgements. The Journal of General Psychology, 129 (3), 257-300.
Verplanken, B. (1993). Need for Cognition and External Information Search:
Responses to Time Pressure during Decision-Making. Journal of Research in Personality, 27, 238-252.
Zautra, A. J., Reich, J. W., Davis, M. C., Potter, P. T. & Nicolson, N .A.
(2000). The Role of Stressful Events in the Relationship Between Positive and Negative Affects: Evidence From Field and Experimental Studies. Journal of Personality, 68, 927-950.
Zuckerman, M. & Kuhlman, D. M. (2000). Personality and Risking Taking:
Common Biosocial Factors. Journal of Personality, 68, 999-1016.
Figure Captions
Figure 1. The Means of Real Math Scores of Functional and Dysfunctional
Impulsive Individuals in a Stressful and Non-Stressful Situation.
Figure 2. The Means of Perceived Math Scores of Functional and Dysfunctional Impulsive in a Stressful and Non-Stressful Situation.
Figure 3. The Means of Real-Perceived Math Scores of Functional and Dysfunctional Impulsive in a Stressful and Non-Stressful Situation.
Appendix
