Effects and Mechanisms of a Deep Breathing Intervention for Test Anxiety — An Exploratory Study on the Use of Mobile EEG Headsets in Educational Research

Project Number
OER 08/15 KKH

Project Duration
June 2015 - May 2018


Test anxiety is a non-trivial issue, especially in high-stakes-examination environments such as the United States, Hong Kong and Singapore. Estimated to afflict 10% to 40% of students from age 7, test anxiety can immediately and adversely impact psychological well-being and performance. In the long run, test anxiety has been linked to low self-esteem, depression and suicide ideation, and is negatively correlated with academic performance and achievement. Although various interventions have been found to reduce test anxiety to varying degrees, the bulk of intervention studies were based on older participants and involved techniques that were relatively complex and required some time “in therapy”. Recognising the need for techniques that can be quickly and easily applied by children, our previous study, funded by an OER start-up grant (SUG), examined the efficacy of a simple deep breathing intervention—a quick and cost-free self-regulatory tool “that can be easily taught and applied in classrooms as young as Grade 2”. Conducted in a school setting, the study found that taking deep breaths before a timed math test significantly reduced feelings of anxiety and improved math performance in elementary school children. However, there were a few areas of ambiguity in the findings that we were unable to clarify with the behavioral data collected. For example, the mechanisms hypothesized to underlie deep breathing’s effect on performance were a better state-of-mind via state anxiety reduction (relaxation) and increased attentional focus. Although state anxiety was found to be reduced by deep breathing, we did not find the expected effect of increased attentional focus. However, positive changes effected by intervention are not always detectable by behavioral measures: Rueda, Checa, and Cómbita (2012) found electroencephalography (EEG) evidence of improved neural efficiency following an attention training intervention, despite the lack of significant improvement on behavioral measures of attention. This suggests that neurophysiological data may provide a more sensitive measure of change—especially in covert cognitive processes—than behavioral data. Furthermore, changes in self-ratings of state anxiety may not reflect actual changes in autonomic arousal (i.e., actual relaxation achieved on a physiological level). The SUG study found some evidence suggesting that dispositional proneness to autonomic reactivity in test-like situations may modulate the effect of deep breathing on performance. Additional neurophysiological data on attention and relaxation may thus be important in elucidating changes effected during the intervention and may provide additional insight on underlying mechanisms. A source of neurophysiological data that has been extensively used in the study of cognitive and affective processes is EEG—non-invasive recordings of brain activity measured by electrodes placed along the scalp. Neurofunctional data measured by techniques such as EEG provide insights on covert cognitive processes which may not always be apparent on a behavioral level. However, the use of EEG outside of laboratory settings—such as, in school-based data collection—has been severely limited by the costs and physical constraints imposed by medical grade EEG systems. Technological advancements in recent years have, however, resulted in new possibilities. Mobile, simplified versions of clinical EEG machines in the form of wireless EEG headsets are now commercially available at relatively low cost. Though originally developed for brain-computer interface (BCI) applications, some of these headsets can also be used to provide EEG-based and raw EEG data. In the proposed study, the use of these headsets to provide additional neurofunctioning data and insight on the effects and possible mechanisms of a deep breathing intervention will be explored.

Funding Source

Related Projects