Search

Emotion regulation habits have long been implicated in risk for depression. However, research in this area traditionally adopts an approach that ignores the multifaceted nature of emotion regulation strategies, the clinical heterogeneity of depression, and potential differential relations between emotion regulation features and individual symptoms. To address limitations associated with the dominant aggregate-level approach, this study aimed to identify which features of key emotion regulation strategies are most predictive and when those features are most predictive of individual symptoms of depression across different time lags. Leveraging novel developments in the field of machine learning, artificial neural network models with feature selection were estimated using data from 460 participants who participated in a 20-wave longitudinal study with weekly assessments. At each wave, participants completed measures of repetitive negative thinking, positive reappraisal, perceived stress, and depression symptoms. Results revealed that specific features of repetitive negative thinking (wondering "why cannot I get going?" and having thoughts or images about feelings of loneliness) and positive reappraisal (looking for positive sides) were important indicators for detecting various depressive symptoms, above and beyond perceived stress. These features had overlapping and unique predictive relations with individual cognitive, affective, and somatic symptoms. Examining temporal fluctuations in the predictive utility, results showed that the utility of these emotion regulation features was stable over time. These findings illuminate potential pathways through which emotion regulation features may confer risk for depression and help to identify actionable targets for its prevention and treatment. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Neocortical sensory areas are generally thought to faithfully represent external stimuli. However, a growing body of evidence suggests that cortical networks exhibit considerable functional plasticity over multiple temporal scales, allowing them to modify their output to reflect ongoing behavioral demands. Nevertheless, the dynamics of sensory and non-sensory representations during acquisition of stimulus-guided behavior are not well understood. We performed longitudinal 2-photon imaging of activity in primary visual cortex (V1) of mice learning a conditioned eyeblink task. We found that both excitatory and inhibitory neurons robustly encode the visual stimulus throughout training despite a significant experience-dependent reduction in response magnitude. In contrast, both pyramidal neurons and parvalbumin-expressing interneurons exhibit emergence of behavioral representation during learning. The plasticity of visual response magnitude and behavioral representation is abolished following loss of NMDA-type glutamate receptors. Overall, our findings demonstrate that V1 networks can dynamically multiplex distinct behaviorally relevant representations over the course of learning.

The goal of voice conversion is to modify a source speaker's speech to sound as if spoken by a target speaker. Common conversion methods are based on Gaussian mixture modeling (GMM). They aim to statistically model the spectral structure of the source and target signals and require relatively large training sets (typically dozens of sentences) to avoid over-fitting. Moreover, they often lead to muffled synthesized output signals, due to excessive smoothing of the spectral envelopes. Mobile applications are characterized with low resources in terms of training data, memory footprint, and computational complexity. As technology advances, computational and memory requirements become less limiting; however, the amount of available training data still presents a great challenge, as a typical mobile user is willing to record himself saying just few sentences. In this paper, we propose the grid-based (GB) conversion method for such low resource environments, which is successfully trained using very few sentences (5---10). The GB approach is based on sequential Bayesian tracking, by which the conversion process is expressed as a sequential estimation problem of tracking the target spectrum based on the observed source spectrum. The converted Mel frequency cepstrum coefficient (MFCC) vectors are sequentially evaluated using a weighted sum of the target training vectors used as grid points. The training process includes simple computations of Euclidian distances between the training vectors and is easily performed even in cases of very small training sets. We use global variance (GV) enhancement to improve the perceived quality of the synthesized signals obtained by the proposed and the GMM-based methods. Using just 10 training sentences, our enhanced GB method leads to converted sentences having closer GV values to those of the target and to lower spectral distances at the same time, compared to enhanced version of the GMM-based conversion method. Furthermore, subjective evaluations show that signals produced by the enhanced GB method are perceived as more similar to the target speaker than the enhanced GMM signals, at the expense of a small degradation in the perceived quality.