It is widely reported that expressive writing can improve mental and physical health. However, to date, the neural correlates of expressive writing have not been reported. The current study examined the neural electrical correlates of expressive writing in a reappraisal approach. Three groups of participants were required to give a public speech. Before speaking, the reappraisal writing group was asked to write about the current stressful task in a reappraisal manner. The irrelevant writing group was asked to write about their weekly plan, and the non-writing group did not write anything. It was found that following the experimental writing manipulation, both reappraisal and irrelevant writing conditions decreased self-reported anxiety levels. But when re-exposed to the stressful situation, participants in the irrelevant writing group showed increased anxiety levels, while anxiety levels remained lower in the reappraisal group. During the experimental writing manipulation period, participants in the reappraisal group had lower frontal alpha asymmetry scores than those in the irrelevant writing group. However, following re-exposure to stress, participants in the reappraisal group showed higher frontal alpha asymmetry scores than those in the irrelevant writing group. Self-reported anxiety and frontal alpha asymmetry of the non-writing condition did not change significantly across these different stages. It is noteworthy that expressive writing in a reappraisal style seems not to be a fast-acting treatment but may instead take effect in the long run.

Emotion recognition based on neural signals is a promising technique for the detection of patients' emotions for enhancing healthcare. However, emotion-related neural signals, such as from functional near infrared spectroscopy (fNIRS), can be affected by various psychophysiological and environmental factors. There is a paucity of literature regarding data instability and classification instability in fNIRS-based emotion recognition systems, phenomenon which may lead to user dissatisfaction and abandonment. We collected data in an fNIRS-based 2-class emotion recognition test-retest experiment (3 week interval) with visual stimuli emotion induction to examine data instability and its impact on classification accuracy. We found a 22.2% average deterioration of emotion classification accuracy between the two sessions, suggesting that classification instability is a serious problem. We found that the changes in the distributions of the selected neural signal features, as evaluated by Kullback-Leibler (KL) divergence, were a likely cause of the accuracy decline. We analyzed the data instability and our results showed that instability of spatial activation patterns and instability of the hemodynamic response in the most activated region are correlated with accuracy decline. Finally, we propose a method for mitigating classification instability in fNIRS-based emotion recognition based on feature selection for stable features, the first such method to our knowledge. This new feature selection criterion considers not only the separability of features (evaluated by Fisher Score) but also their stability over time (evaluated by KL divergence between feature distributions at different time points). Testing showed that this method led to an approximately 5% improvement in cross-session generalization accuracy.

Functional near-infrared spectroscopy (fNIRS) is being increasingly applied to affective and social neuroscience research; however, the reliability of this method is still unclear. This study aimed to evaluate the test–retest reliability of the fNIRS-based prefrontal response to emotional stimuli. Twenty-six participants viewed unpleasant and neutral pictures, and were simultaneously scanned by fNIRS in two sessions three weeks apart. The reproducibility of the prefrontal activation map was evaluated at three spatial scales (mapwise, clusterwise, and channelwise) at both the group and individual levels. The influence of the time interval was also explored and comparisons were made between longer (intersession) and shorter (intrasession) time intervals. The reliabilities of the activation map at the group level for the mapwise (up to 0.88, the highest value appeared in the intersession assessment) and clusterwise scales (up to 0.91, the highest appeared in the intrasession assessment) were acceptable, indicating that fNIRS may be a reliable tool for emotion studies, especially for a group analysis and under larger spatial scales. However, it should be noted that the individual-level and the channelwise fNIRS prefrontal responses were not sufficiently stable. Future studies should investigate which factors influence reliability, as well as the validity of fNIRS used in emotion studies.

When irritated by other people, powerful people usually tend to express their anger explicitly and directly, whereas people in less powerful positions are more likely not to show their feelings freely. The neural mechanism behind power and its influence on expression tendency has been scarcely explored. This study recorded frontal EEG activity at rest and frontal EEG activation while participants were engaged in a writing task describing an anger-eliciting event, in which they were irritated by people with higher or lower social power. Participants’ anger levels and expression inclination levels were self-reported on nine-point visual analog Likert scales, and also rated by independent raters based on the essays they had written. The results showed that high social power was indeed associated with greater anger expression tendency and greater left frontal activation than low social power. This is in line with the approach-inhibition theory of power. The mid-frontal asymmetric activation served as a partial mediator between social power and expression inclination. This effect may relate to the functions of the prefrontal cortex, which is in charge of information integration and evaluation and the control of motivation direction, as reported by previous studies.