It is of significant importance to detect and manage stress before it turns into severe problems. However, existing stress detection methods usually rely on psychological scales or physiological devices, making the detection complicated and costly. In this paper, we explore to automatically detect individuals' psychological stress via social media. Employing real online micro-blog data, we first investigate the correlations between users' stress and their tweeting content, social engagement and behavior patterns. Then we define two types of stress-related attributes: 1) low-level content attributes from a single tweet, including text, images and social interactions; 2) user-scope statistical attributes through their weekly micro-blog postings, leveraging information of tweeting time, tweeting types and linguistic styles. To combine content attributes with statistical attributes, we further design a convolutional neural network (CNN) with cross autoencoders to generate user-scope content attributes from low-level content attributes. Finally, we propose a deep neural network (DNN) model to incorporate the two types of userscope attributes to detect users' psychological stress. We test the trained model on four different datasets from major micro-blog platforms including Sina Weibo, Tencent Weibo and Twitter. Experimental results show that the proposed model is effective and efficient on detecting psychological stress from micro-blog data. We believe our model would be useful in developing stress detection tools for mental health agencies and individuals.

No abstract

The direct modulation of subthalamic nucleus (STN) neurons by dopamine (DA) neurons of the substantia nigra (SN) is controversial owing to the thick caliber and low density of DA axons in the STN. The abnormal activity of the STN in Parkinson's disease (PD), which is central to the appearance of symptoms, is therefore thought to result from the loss of DA in the striatum. We carried out three experiments in rats to explore the function of DA in the STN: (i) light and electron microscopic analysis of tyrosine hydroxylase (TH)-, dopamine beta-hydroxylase (DbetaH)- and DA-immunoreactive structures to determine whether DA axons form synapses; (ii) fast-scan cyclic voltammetry (FCV) to determine whether DA axons release DA; and (iii) patch clamp recording to determine whether DA, at a concentration similar to that detected by FCV, can modulate activity and synaptic transmission/integration. TH- and DA-immunoreactive axons mostly formed symmetric synapses. Because DbetaH-immunoreactive axons were rare and formed asymmetric synapses, they comprised the minority of TH-immunoreactive synapses. Voltammetry demonstrated that DA release was sufficient for the activation of receptors and abolished by blockade of voltage-dependent Na+ channels or removal of extracellular Ca2+. The lifetime and concentration of extracellular DA was increased by blockade of the DA transporter. Dopamine application depolarized STN neurons, increased their frequency of activity and reduced the impact of gamma-aminobutyric acid (GABA)-ergic inputs. These findings suggest that SN DA neurons directly modulate the activity of STN neurons and their loss may contribute to the abnormal activity of STN neurons in PD.

A recent study reported lower anxiety in the 5xFAD transgenic mouse model of Alzheimer's disease, as measured by reduced time on the open arms of an elevated plus maze. This is important because all behaviors in experimental animals must be interpreted in light of basal anxiety and response to novel environments. We conducted a comprehensive anxiety battery in the 5xFAD transgenics and replicated the plus-maze phenotype. However, we found that it did not reflect reduced anxiety, but rather abnormal avoidance of the closed arms on the part of transgenics and within-session habituation to the closed arms on the part of wild-type controls. We noticed that the 5xFAD transgenics did not engage in the whisker-barbering behavior typical of mice of this background strain. This is suggestive of abnormal social behavior, and we suspected it might be related to their avoidance of the closed arms on the plus maze. Indeed, transgenic mice exhibited excessive home-cage social behavior and impaired social recognition, and did not permit barbering by wild-type mice when pair-housed. When their whiskers were snipped the 5xFAD transgenics no longer avoided the closed arms on the plus maze. Examination of parvalbumin (PV) staining showed a 28.9% reduction in PV+ inhibitory interneurons in the in barrel fields of 5xFAD mice, and loss of PV+ fibers in layers IV and V. This loss of vibrissal inhibition suggests a putatively aversive overstimulation that may be responsible for the transgenics’ avoidance of the closed arms in the plus maze.