scholarly journals How the Discrepancy Between Prior Expectations and New Information Influences Expectation Updating in Depression—The Greater, the Better?

2021 ◽  
pp. 216770262110246
Author(s):  
Tobias Kube ◽  
Lukas Kirchner ◽  
Gunnar Lemmer ◽  
Julia Anna Glombiewski
Keyword(s):  
Social Interaction ◽  
Prediction Error ◽  
Tipping Point ◽  
Small Study ◽  
Traditional Learning ◽  
Learning Models ◽  
False Feedback ◽  
New Information ◽  
Interaction Task ◽  
Disconfirming Information

Previous research on expectation updating in relation to psychopathology used to treat expectation-confirming information and expectation-disconfirming information as binary concepts. Here, we varied the extent to which new information deviates from prior expectations and examined its influence on expectation adjustment in both a false-feedback task (Study 1; N = 379) and a social-interaction task (Study 2; N = 292). Unlike traditional learning models, we hypothesized a tipping point in which the discrepancy between expectation and outcome becomes so large that new information is perceived as lacking credibility, thus entailing little updating of expectations. Consistent with the hypothesized tipping point, new information was deemed most valid if it was moderately positive. Moreover, descriptively, expectation update was largest for moderate expectation violations, but this effect was small (Study 2) or even nonsignificant (Study 1). The findings question the assumption of traditional learning models that the larger the prediction error, the larger the update.

How the positivity of new information influences belief updating in depression - The more, the better?

2021 ◽  
Author(s):  
Tobias Kube ◽  
Lukas Kirchner ◽  
Gunnar Lemmer ◽  
Julia Glombiewski
Keyword(s):  
Depressive Symptoms ◽  
Traditional Learning ◽  
False Feedback ◽  
Prior Beliefs ◽  
Belief Updating ◽  
Positive Information ◽  
New Information ◽  
New Learning ◽  
Interaction Task ◽  
Disconfirming Information

Aberrant belief updating has been linked to psychopathology, e.g., depressive symptoms. While previous research used to treat belief-confirming vs. -disconfirming information as binary concepts, the present research varied the extent to which new information deviates from prior beliefs and examined its influence on belief updating. In a false feedback task (Study 1; N = 379) and a social interaction task (Study 2; N = 292), participants received slightly positive, moderately positive or extremely positive information in relation to their prior beliefs. In both studies, new information was deemed most reliable if it was moderately positive. Yet, differences in the positivity of new information had only small effects on belief updating. In Study 1, depressive symptoms were related to difficulties in generalizing positive new learning experiences. The findings suggest that, contrary to traditional learning models, the larger the differences between prior beliefs and new information, the more beliefs are not updated.

Factors influencing the update of beliefs regarding controversial political issues

2022 ◽  
Author(s):  
Tobias Kube
Keyword(s):  
Prediction Error ◽  
Public Discourse ◽  
Prior Beliefs ◽  
Belief Updating ◽  
Main Hypothesis ◽  
Political Issues ◽  
New Information ◽  
The World ◽  
Disconfirming Information ◽  
Selective Integration

When updating beliefs in light of new information, people preferentially integrate information that is consistent with their prior beliefs and helps them construe a coherent view of the world. Such a selective integration of new information likely contributes to belief polarisation and compromises public discourse. Therefore, it is crucial to understand the factors that underlie biased belief updating. To this end, I conducted three pre-registered experiments covering different controversial political issues (i.e., Experiment 1: climate change, Experiment 2: speed limit on highways, Experiment 3: immigration in relation to violent crime). The main hypothesis was that negative reappraisal of new information (referred to as “cognitive immunisation”) hinders belief updating. Support for this hypothesis was found only in Experiment 2. In all experiments, the magnitude of the prediction error (i.e., the discrepancy between prior beliefs and new information) was strongly related to belief updating. Across experiments, participants’ general attitudes regarding the respective issue influenced the strength of beliefs, but not their update. The present findings provide some indication that the engagement in cognitive immunisation can lead to the maintenance of beliefs despite disconfirming information. However, by far the largest association with belief updating was with the magnitude of the prediction error.

Framework for TCAD augmented machine learning on multi- I–V characteristics using convolutional neural network and multiprocessing

2021 ◽  
Vol 42 (12) ◽  
pp. 124101
Author(s):  
Thomas Hirtz ◽  
Steyn Huurman ◽  
He Tian ◽  
Yi Yang ◽  
Tian-Ling Ren
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Information Technologies ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Data Driven ◽  
Sufficient Data ◽  
Learning Models ◽  
Simulation Tools ◽  
New Information

Abstract In a world where data is increasingly important for making breakthroughs, microelectronics is a field where data is sparse and hard to acquire. Only a few entities have the infrastructure that is required to automate the fabrication and testing of semiconductor devices. This infrastructure is crucial for generating sufficient data for the use of new information technologies. This situation generates a cleavage between most of the researchers and the industry. To address this issue, this paper will introduce a widely applicable approach for creating custom datasets using simulation tools and parallel computing. The multi-I–V curves that we obtained were processed simultaneously using convolutional neural networks, which gave us the ability to predict a full set of device characteristics with a single inference. We prove the potential of this approach through two concrete examples of useful deep learning models that were trained using the generated data. We believe that this work can act as a bridge between the state-of-the-art of data-driven methods and more classical semiconductor research, such as device engineering, yield engineering or process monitoring. Moreover, this research gives the opportunity to anybody to start experimenting with deep neural networks and machine learning in the field of microelectronics, without the need for expensive experimentation infrastructure.

scholarly journals Deep Transfer Learning for Source Code Modeling

2020 ◽  
Vol 30 (05) ◽  
pp. 649-668
Author(s):  
Yasir Hussain ◽  
Zhiqiu Huang ◽  
Yu Zhou ◽  
Senzhang Wang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Transfer Learning ◽  
Related Problem ◽  
State Of The Art ◽  
Source Code ◽  
Traditional Learning ◽  
Learning Models ◽  
Modeling And Analysis ◽  
F Measure

In recent years, deep learning models have shown great potential in source code modeling and analysis. Generally, deep learning-based approaches are problem-specific and data-hungry. A challenging issue of these approaches is that they require training from scratch for a different related problem. In this work, we propose a transfer learning-based approach that significantly improves the performance of deep learning-based source code models. In contrast to traditional learning paradigms, transfer learning can transfer the knowledge learned in solving one problem into another related problem. First, we present two recurrent neural network-based models RNN and GRU for the purpose of transfer learning in the domain of source code modeling. Next, via transfer learning, these pre-trained (RNN and GRU) models are used as feature extractors. Then, these extracted features are combined into attention learner for different downstream tasks. The attention learner leverages from the learned knowledge of pre-trained models and fine-tunes them for a specific downstream task. We evaluate the performance of the proposed approach with extensive experiments with the source code suggestion task. The results indicate that the proposed approach outperforms the state-of-the-art models in terms of accuracy, precision, recall and F-measure without training the models from scratch.

scholarly journals Opinion: Neural differentiation of incorrectly predicted memories

2018 ◽  
Author(s):  
Andrea Greve ◽  
Hunar Abdulrahman ◽  
Richard Henson
Keyword(s):  
Prediction Error ◽  
Neural Differentiation ◽  
Recent Article ◽  
Transitive Inference ◽  
Experimental Conditions ◽  
Neural Representations ◽  
New Information

In their recent article ‘Neural differentiation of incorrectly predicted memories’, Kim et al. (2017) investigate how neural representations of items change when they are incorrectly predicted and subsequently restudied. The authors conclude such items undergo representational differentiation, i.e. a decreased overlap in the representations of an item and its context. We suggest the results are equally compatible with the reverse mechanism of integration, i.e. increased learning of new information and present simulations to demonstrate this. More importantly, we show how new experimental conditions could distinguish integration from differentiation and discuss how the results fit with recent suggestions about prediction-error driven learning and transitive inference.

scholarly journals Improvement of Learning Outcomes Through The Solution to Apply Lack of Student Social Interaction with The Application of Discovery Learning Models in SD Negeri 1 Pagedangan

2020 ◽  
Vol 3 (3) ◽  
pp. 1280
Author(s):  
Emilia Sri Dwi Lestari
Keyword(s):  
Elementary Schools ◽  
Social Interaction ◽  
Action Research ◽  
Learning Outcomes ◽  
Student Learning Outcomes ◽  
Discovery Learning ◽  
Learning Models ◽  
Audiovisual Media ◽  
Post Test ◽  
Improve Student Learning

<pre><em>The purpose of this study is to improve learning outcomes through a solution to the application of the lack of social interaction of students in grade IV elementary schools with the audiovisual media assisted Discovery Learning model. The research conducted was a Classroom Action Research (CAR) in two cycles, with each cycle consisting of one meeting. The stages of each cycle are planning, implementing, observing and reflecting. Each meeting is carried out a pre test and post test to determine the progress of students. In the first cycle students who completed after carrying out the post test were 58%. In the second cycle students who completed after carrying out the post test were 75%. These results indicate that the application of audiovisual media can improve student learning outcomes, especially Indonesian language muple at SD Negeri 1 Pagedangan.</em><em></em></pre>

Connecting Web 2.0 to Connectivism in Supporting Students' Learning

2013 ◽  
Vol 1 (2) ◽  
pp. 135-145
Author(s):  
Siti Ilyana Mohd Yusof ◽  
Nor Hasbiah Ubaidullah ◽  
Zulkifley Mohamed
Keyword(s):  
Social Interaction ◽  
Web 2.0 ◽  
Communication Networks ◽  
Social Networking Sites ◽  
Learning Theories ◽  
Traditional Learning ◽  
Personal Learning ◽  
Information Sets ◽  
Applicable Knowledge ◽  
Fundamental Insight

Objective- Students' learning has been transformed by the advent of Web 2.0 which is defined as more personalized and a communicative form of the World Wide Web. This paper is positioned within the context of Web 2.0 through connectivism in changing the educational environment. Methodology/Technique Connectivism is a learning theory of the digital age, which reflects social interaction as part of the learning process. In contrast to traditional theories, students' learning can result from social interaction. Students' learning can visualised as connectivity; people derive skills and competencies from forming connections while focusing on connecting specialised information sets. Connectivity has established communication networks that enable students to obtain applicable knowledge and experiences. Findings Web 2.0 tools such as blogs, social networking sites and wikis allow for a variety of online social interactions and moulding the way people relate to each other. It also supports students' learning through the lens of connectivism. Novelty - People can still learn by applying the traditional learning theories, but the fundamental insight, aligning with the underpinning connectivism, relates to people's ability to construct their own social networks that integrates with their personal learning environments to foster and sustain the flow of knowledge. Type of Paper Empirical paper Keywords: , Web 2.0; connectivism; students' learning

Elaborated Delusional Beliefs

2020 ◽  
pp. 70-92
Author(s):  
Lisa Bortolotti
Keyword(s):  
Cognitive Performance ◽  
Self Efficacy ◽  
Prediction Error ◽  
Error Theory ◽  
Automated Learning ◽  
New Information ◽  
Previous State ◽  
Epistemic Innocence

In this chapter, the author argues that delusional beliefs that are elaborated—often emerging in people who attract a diagnosis of schizophrenia—have the potential for epistemic innocence. Delusional beliefs are strenuously resistant to counterevidence. However, when they are adopted to explain a puzzling experience that might compromise the agents’ capacity to interact with their environment, delusional beliefs contribute to restoring some aspects of cognitive performance by temporarily reducing anxiety. On the prediction-error theory of delusion formation, it is further believed that the adoption of a delusional explanation helps resume the processes of automated learning compromised by inaccurate prediction-error signalling. Depending on their content, some delusional beliefs may also support an attitude of curiosity and self-efficacy that is more conducive to the acquisition of new information than the previous state of uncertainty and self-doubt.

Litter Size, Emotionality, Learning, and Social Interaction in C57Bl/6J Mice

1976 ◽  
Vol 39 (1) ◽  
pp. 27-31 ◽  
Author(s):  
Robert Deitchman ◽  
Alan Lavine ◽  
Joel Burkholder
Keyword(s):  
Social Interaction ◽  
Open Field ◽  
Litter Size ◽  
Maternal Behavior ◽  
Body Contact ◽  
The Social ◽  
Escape Learning ◽  
Interaction Task ◽  
Preweaning Period

Activity in the open field, weight, escape learning, and social interaction were assessed as a function of amount of stimulation during the preweaning period in C57BL/6J mice. The results showed that subjects in small litters (5 pups) were more active and had shorter latencies to cross the first square in the open field than subjects in large litters (10 pups). The social-interaction task indicated that subjects in large litters had shorter latencies to initiate body contact. However, no differences in weight or escape learning were reported. Competition for nipple sites, the amount of maternal behavior, and expectancy of contact were described as possible variables affecting the performance of subjects reared in large and small litters.

Sign in / Sign up

Export Citation Format

Share Document