You are here: Home / The IPN / Archive / Self-control and Cognitive Fatigue in Test Situations

Self-control and Cognitive Fatigue in Test Situations

April 15th, 2018

Cognitive fatigue during a performance test has a negative effect on test results


Christoph Lindner


Self-control is the ability to suppress impulses (e.g., thoughts, emotions), regulate attention, and modify prevalent behavioral tendencies to achieve higher goals. In addition to the basic cognitive abilities, self-control is regarded as an evolutionary function and most important prerequisite for efficient behavioral adaptation and thus enables the execution of purposeful actions. The exercise of self-control is therefore subject to the conscious will. In literature, self-control is often used synonymously with the terms self-discipline or self-regulation.

Ego-Depletion Effect
People are not self-controlled in all situations. Successful self-control is assumed to be subject to a limited mental capacity, exhausted in the short term by the use of self-control over time. This mental exhaustion effect is referred to as ego depletion. Behavioral experiments are carried out in basic psychological research to investigate ego-depletion effects. Compared to a control group, subjects perform worse in self-control-intensive tasks that require a high degree of attention focus or logical thinking, for example, if they had to use self-control immediately before and are therefore in the ego-depleted state.

Dispositional self-control
Dispositional self-control is the temporarily relatively stable proportion of self-control behavior which can be described as an individual personality trait. People with high levels of dispositional self-control tend to behave more conscientiously than people with low dispositional self-control. In addition, they are less distracted and prove more persistent and motivated in learning and achievement situations. Beyond the basic cognitive abilities, dispositional self-control proves to be the strongest predictor for predicting the performance behavior of students.


To successfully take achievement tests, test participants must access subject-specific knowledge and use self-control to focus their attention on the content of the task without being distracted. Numerous findings from basic research indicate that the longer-term exercise of self-control can lead to a state of mental exhaustion (ego-depletion effect), which temporarily impairs the targeted regulation of thought and attention. Consequences of such ego-depletion effects include test takers exhibiting poorer performance in working memory tests or in cognitive tests for logical reasoning. Since solving mathematical and scientific achievement tests in particular requires targeted attention and thought control as well as logical reasoning, ego-depletion effects could occur during test taking and have a negative effect on test behavior, performance and thus also on the validity of competence measurements, irrespective of the competence of the test participants. Recent studies, however, report that people with a high degree of dispositional self-control may be even more affected by ego-depletion effects and thus have a weaker performance. Based on these contradictory findings, the aim of this work was to investigate the development and effect of ego-depletion effects in greater detail taking into account dispositional self-control in the context of mathematical-scientific school achievement tests. To investigate the relationships outlined above, one must ensure that the dispositional self-control is validly assessed in the context of performance. It has not yet been clarified to which extent the typically used instrument, the Brief Self-Control Scale (BSCS), actually represents the dispositional self-control validly and reliably. Despite the wide range of BSCS applications, there is no clear empirical evidence for its theoretically assumed one-dimensional structure, which is why different multidimensional conceptualizations of BSCS have been suggested in the literature. So far it is unknown to what extent these multidimensional scale versions of the one-dimensional BSCS are superior to predicting performance behavior in connection with pedagogical-psychological questions. The main concerns of the present studies were thus comparing different BSCS conceptualizations for measuring dispositional self-control in the context of performance (Study 1) as well as to examine the development (Study 2) and effect (Study 3) of ego-deposition effects in test situations taking dispositional self-control into account in a differentiated manner. For the implementation of the research project, two field studies (studies 1 and 2) were carried out in the project "Mathematical and scientific competences in initial vocational training (ManKobE)" as well as a behavioral experiment (Study 3) funded by the Leibniz Association (in the framework of the Leibniz Competition by the Senate Competition Committee).

The first study compared the validity of different conceptualizations of BSCS. Traditionally, when using BSCS, the mean value over all items is used to represent self-control as a one-dimensional construct. However, several authors exclude items from the overall scale and use the mean value of different item groupings to represent different facets of dispositional self-control. Using two large samples of students (N = 205) and trainees (N = 1951), the factor structures of different BSCS conceptualizations in both samples were investigated by means of confirmatory factor analysis. Furthermore, correlations with outcome variables for the BSCS's total score, the facet scores and for each two-dimensional conceptualization (i.e., multiple correlations) were calculated. Although the internal structure of the BSCS could not be confirmed as one-dimensional, as assumed, the overall results show that the use of the scales’ mean value is the best solution for predicting achievement-related outcome variables. .

The second study examined the extent to which students’ state self-control capacity as an indicator of the extent of the ego-depletion effect (the lower the perceived self-control capacity, the greater the exhaustion from ego-depletion) and their test-taking effort changed simultaneously over the course of a school achievement test, while taking dispositional self-control into account. Therefore, latent growth curve modelling was applied to investigate temporal changes of students’ state self-control capacity and effort-investment. As assumed, there was a correlation between the decrease in the state self-control capacity and the decreasing effort invested during testing. Overall, individuals with high dispositional self-control put in more effort and had a higher average state self-control capacity throughout the test, but experienced a stronger decline than those with low dispositional self-control.

The third study examined the effect of an experimentally induced ego-depletion effect on processing time and performance in a computer-based mathematical problem-solving test (45 tasks), taking dispositional self-control into account. For the implementation, 120 students from comprehensive schools in Schleswig-Holstein were randomly assigned to an experimental group and a control group. For the manipulation of ego depletion, the experimental group had to use self-control to resist the impulse to completely copy words when transcribing a text and instead omit all "e" and "n" letters. The control group did not have to use any self-control and simply copied the identical text. Afterwards all participants worked on the mathematics test. Using logistic multi-level regression analyses, the progressions of task completion times and solution probabilities for the experimental group compared to the control group were modelled and the effects of dispositional self-control were also taken into account.

As expected, over the course of the test the mathematical performance decreased significantly as a result of the ego depletion intervention. After about half of the test, the experimental group was significantly less likely to solve the tasks than the control group. In addition, test participants with high levels of dispositional self-control on average invested more effort and time in completing their mathematical items. Interestingly, people with high dispositional self-control in the ego depletion state showed a greater reduction in mathematical performance over the course of the test. This reflects the paradoxical effect reported in the literature that those with high levels of dispositional self-control are more affected by ego-depletion effects than those with low dispositional self-control.

In summary, these findings indicate that self-control plays an important role in performance tests and achievement situations. As expected, the development and effect of ego depletion could be proven in the context of school performance tests. Modeling self-control dependent changes of psychological variables made it possible to identify a previously unnoticed interplay between the dispositional self-control and the changes in students’ test-taking effort and state self-control capacity. These results contribute to a better understanding of self-controlled behavior in achievement situations.