scholarly journals Model-based aversive learning in humans is supported by preferential task state reactivation

2021 ◽  
Vol 7 (31) ◽  
pp. eabf9616
Author(s):  
Toby Wise ◽  
Yunzhe Liu ◽  
Fatima Chowdhury ◽  
Raymond J. Dolan
Keyword(s):  
Mental Model ◽  
Learning Task ◽  
Harm Avoidance ◽  
Neural Mechanisms ◽  
Trial And Error ◽  
Aversive Learning ◽  
Theta Power ◽  
Model Based ◽  
Outcome Valence ◽  
Hippocampal Theta

Harm avoidance is critical for survival, yet little is known regarding the neural mechanisms supporting avoidance in the absence of trial-and-error experience. Flexible avoidance may be supported by a mental model (i.e., model-based), a process for which neural reactivation and sequential replay have emerged as candidate mechanisms. During an aversive learning task, combined with magnetoencephalography, we show prospective and retrospective reactivation during planning and learning, respectively, coupled to evidence for sequential replay. Specifically, when individuals plan in an aversive context, we find preferential reactivation of subsequently chosen goal states. Stronger reactivation is associated with greater hippocampal theta power. At outcome receipt, unchosen goal states are reactivated regardless of outcome valence. Replay of paths leading to goal states was modulated by outcome valence, with aversive outcomes associated with stronger reverse replay than safe outcomes. Our findings are suggestive of avoidance involving simulation of unexperienced states through hippocampally mediated reactivation and replay.

scholarly journals Model-based aversive learning in humans is supported by preferential task state reactivation

2020 ◽  
Author(s):  
Toby Wise ◽  
Yunzhe Liu ◽  
Fatima Chowdhury ◽  
Raymond J. Dolan
Keyword(s):  
Learning Task ◽  
Harm Avoidance ◽  
Neural Mechanisms ◽  
Avoidance Behaviour ◽  
Potential Candidate ◽  
Trial And Error ◽  
Aversive Learning ◽  
Outcome Valence ◽  
Alternative Future ◽  
Hippocampal Theta

AbstractHarm avoidance is critical for survival, yet little is known regarding the underlying neural mechanisms supporting avoidance when we cannot rely on direct trial and error experience. Neural reactivation, and sequential replay, have emerged as potential candidate mechanisms. Here, during an aversive learning task, in conjunction with magnetoencephalography, we show prospective and retrospective reactivation for planning and learning respectively, coupled to evidence for sequential replay. Specifically, when subjects plan in an aversive context, we find preferential reactivation of subsequently chosen goal states and sequential replay of the preceding path. This reactivation was associated with greater hippocampal theta power. At outcome receipt, unchosen goal states are reactivated regardless of outcome valence. However, replay of paths leading to goal states was directionally modulated by outcome valence, with aversive outcomes leading to stronger reverse replay compared to safe outcomes. Our findings suggest that avoidance behaviour involves simulation of alternative future and past outcome states through hippocampally-mediated reactivation and replay.

scholarly journals Dynamic integration of conceptual information during learning

2018 ◽  
Author(s):  
Marika C. Inhoff ◽  
Laura A. Libby ◽  
Takao Noguchi ◽  
Bradley C. Love ◽  
Charan Ranganath
Keyword(s):  
Computational Model ◽  
Learning Task ◽  
Higher Order ◽  
Neural Mechanisms ◽  
Semantic Features ◽  
Learning Problem ◽  
Conceptual Information ◽  
Conceptual Representations ◽  
Model Based ◽  
Learning Rates

AbstractThe development and application of concepts is a critical component of cognition. Although concepts can be formed on the basis of simple perceptual or semantic features, conceptual representations can also capitalize on similarities across feature relationships. By representing these types of higher-order relationships, concepts can simplify the learning problem and facilitate decisions. Despite this, little is known about the neural mechanisms that support the construction and deployment of these kinds of higher-order concepts during learning. To address this question, we combined a carefully designed associative learning task with computational model-based functional magnetic resonance imaging (fMRI). Participants were scanned as they learned and made decisions about sixteen pairs of cues and associated outcomes. Associations were structured such that individual cues shared feature relationships, operationalized as shared patterns of cue pair-outcome associations. In order to capture the large number of possible conceptual representational structures that participants might employ and to evaluate how conceptual representations are used during learning, we leveraged a well-specified Bayesian computational model of category learning [1]. Behavioral and model-based results revealed that participants who displayed a tendency to link experiences in memory benefitted from faster learning rates, suggesting that the use of the conceptual structure in the task facilitated decisions about cue pair-outcome associations. Model-based fMRI analyses revealed that trial-by-trial integration of cue information into higher-order conceptual representations was supported by an anterior temporal (AT) network of regions previously implicated in representing complex conjunctions of features and meaning-based information.

Model-Based and Model-Free Social Cognition

2019 ◽  
Author(s):  
Leor M Hackel ◽  
Jeffrey Jordan Berg ◽  
Björn Lindström ◽  
David Amodio
Keyword(s):  
Reinforcement Learning ◽  
Social Cognition ◽  
Learning Strategies ◽  
Memory Systems ◽  
Learning Task ◽  
Financial Advisors ◽  
Model Based ◽  
Model Free ◽  
Systems Model ◽  
Task Assessment

Do habits play a role in our social impressions? To investigate the contribution of habits to the formation of social attitudes, we examined the roles of model-free and model-based reinforcement learning in social interactions—computations linked in past work to habit and planning, respectively. Participants in this study learned about novel individuals in a sequential reinforcement learning paradigm, choosing financial advisors who led them to high- or low-paying stocks. Results indicated that participants relied on both model-based and model-free learning, such that each independently predicted choice during the learning task and self-reported liking in a post-task assessment. Specifically, participants liked advisors who could provide large future rewards as well as advisors who had provided them with large rewards in the past. Moreover, participants varied in their use of model-based and model-free learning strategies, and this individual difference influenced the way in which learning related to self-reported attitudes: among participants who relied more on model-free learning, model-free social learning related more to post-task attitudes. We discuss implications for attitudes, trait impressions, and social behavior, as well as the role of habits in a memory systems model of social cognition.

scholarly journals Sentiment Analysis on Twitter Airline Data

2021 ◽  
Vol 9 (VI) ◽  
pp. 3767-3770
Author(s):  
Kirti Jain
Keyword(s):  
Machine Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Sentiment Analysis ◽  
Language Processing ◽  
Learning Task ◽  
Model Based ◽  
Sentiment Mining ◽  
General Opinion

Sentiment analysis, also known as sentiment mining, is a submachine learning task where we want to determine the overall sentiment of a particular document. With machine learning and natural language processing (NLP), we can extract the information of a text and try to classify it as positive, neutral, or negative according to its polarity. In this project, We are trying to classify Twitter tweets into positive, negative, and neutral sentiments by building a model based on probabilities. Twitter is a blogging website where people can quickly and spontaneously share their feelings by sending tweets limited to 140 characters. Because of its use of Twitter, it is a perfect source of data to get the latest general opinion on anything.

scholarly journals Learning to synchronize: Midfrontal theta dynamics during rule switching

2020 ◽  
Author(s):  
Pieter Verbeke ◽  
Kate Ergo ◽  
Esther De Loof ◽  
Tom Verguts
Keyword(s):  
Negative Feedback ◽  
Human Subjects ◽  
Learning Task ◽  
Theta Oscillations ◽  
Prediction Errors ◽  
Hierarchical Learning ◽  
Theta Power ◽  
Negative Prediction ◽  
Theta Phase ◽  
The Brain

AbstractIn recent years, several hierarchical extensions of well-known learning algorithms have been proposed. For example, when stimulus-action mappings vary across time or context, the brain may learn two or more stimulus-action mappings in separate modules, and additionally (at a hierarchically higher level) learn to appropriately switch between those modules. However, how the brain mechanistically coordinates neural communication to implement such hierarchical learning, remains unknown. Therefore, the current study tests a recent computational model that proposed how midfrontal theta oscillations implement such hierarchical learning via the principle of binding by synchrony (Sync model). More specifically, the Sync model employs bursts at theta frequency to flexibly bind appropriate task modules by synchrony. 64-channel EEG signal was recorded while 27 human subjects (Female: 21, Male: 6) performed a probabilistic reversal learning task. In line with the Sync model, post-feedback theta power showed a linear relationship with negative prediction errors, but not with positive prediction errors. This relationship was especially pronounced for subjects with better behavioral fit (measured via AIC) of the Sync model. Also consistent with Sync model simulations, theta phase-coupling between midfrontal electrodes and temporo-parietal electrodes was stronger after negative feedback. Our data suggest that the brain uses theta power and synchronization for flexibly switching between task rule modules, as is useful for example when multiple stimulus-action mappings must be retained and used.Significance StatementEveryday life requires flexibility in switching between several rules. A key question in understanding this ability is how the brain mechanistically coordinates such switches. The current study tests a recent computational framework (Sync model) that proposed how midfrontal theta oscillations coordinate activity in hierarchically lower task-related areas. In line with predictions of this Sync model, midfrontal theta power was stronger when rule switches were most likely (strong negative prediction error), especially in subjects who obtained a better model fit. Additionally, also theta phase connectivity between midfrontal and task-related areas was increased after negative feedback. Thus, the data provided support for the hypothesis that the brain uses theta power and synchronization for flexibly switching between rules.

Discriminative aversive learning and amygdala responsivity is enhanced in mice with reduced serotonin transporter activity

2017 ◽  
Author(s):  
João Lima ◽  
Trevor Sharp ◽  
Amy M. Taylor ◽  
David M. Bannerman ◽  
Stephen B. McHugh
Keyword(s):  
Discrimination Learning ◽  
Learning Task ◽  
Plasticity Theory ◽  
Fear Learning ◽  
Aversive Learning ◽  
Hemodynamic Responses ◽  
Knock Out ◽  
Auditory Cue ◽  
Network Plasticity ◽  
Knock Out Mice

AbstractThe serotonin (5-HT) transporter (5-HTT) regulates 5-HT availability at the synapse. Low or null 5-HTT expression results in increased 5-HT availability and has been reported to produce anxious and depressive phenotypes, although this remains highly controversial despite two decades of investigation. Paradoxically, SSRIs, which also increase 5-HT availability, reduce the symptoms of anxiety and depression. An emerging ‘network plasticity’ theory of 5-HT function argues that, rather than influencing mood directly, increasing 5-HT availability enhances learning about emotionally-significant events but evidence supporting this theory is inconclusive. Here, we tested one key prediction of this theory: that increased 5-HT availability enhances aversive learning. In experiment 1, we trained 5-HTT knock-out mice (5-HTTKO), which have increased 5-HT availability, and wild-type mice (WT) on an aversive discrimination learning task in which one auditory cue was paired with an aversive outcome whereas a second auditory cue was not. Simultaneously we recorded neuronal and hemodynamic responses from the amygdala, a brain region necessary for aversive learning. 5-HTTKO mice exhibited superior discrimination learning than WTs, and had stronger theta-frequency neuronal oscillations and larger amygdala hemodynamic responses to the aversive cues, which predicted the extent of learning. In experiment 2, we found that acute SSRI treatment (in naïve non-transgenic mice), given specifically before fear learning sessions, enhanced subsequent fear memory recall. Collectively, our data demonstrate that reducing 5-HTT activity (and thereby increasing 5-HT availability) enhances amygdala responsivity to aversive events and facilitates learning for emotionally-relevant cues. Our findings support the network plasticity theory of 5-HT function.

Theta power and theta-gamma coupling support spatial memory retrieval

2019 ◽  
Author(s):  
Umesh Vivekananda ◽  
Daniel Bush ◽  
James A Bisby ◽  
Sallie Baxendale ◽  
Roman Rodionov ◽  
...  
Keyword(s):  
Spatial Memory ◽  
Task Performance ◽  
Memory Function ◽  
Memory Task ◽  
Theta Oscillations ◽  
Theta Power ◽  
Human Epilepsy ◽  
Phase Amplitude ◽  
Hippocampal Theta

AbstractHippocampal theta oscillations have been implicated in spatial memory function in both rodents and humans. What is less clear is how hippocampal theta interacts with higher frequency oscillations during spatial memory function, and how this relates to subsequent behaviour. Here we asked ten human epilepsy patients undergoing intracranial EEG recording to perform a desk-top virtual reality spatial memory task, and found that increased theta power in two discrete bands (‘low’ 2-5Hz and ‘high’ 6-9Hz) during cued retrieval was associated with improved task performance. Similarly, increased coupling between ‘low’ theta phase and gamma amplitude during the same period was associated with improved task performance. These results support a role of theta oscillations and theta-gamma phase-amplitude coupling in human spatial memory function.

Serotonergic modulation of cognition; An acute challenge

2017 ◽  
Vol 41 (S1) ◽  
pp. S350-S350
Author(s):  
N. Skandali ◽  
J. Rowe ◽  
J. Deakin ◽  
T. Robbins ◽  
B. Sahakian
Keyword(s):  
Instrumental Learning ◽  
Learning Task ◽  
Treated Group ◽  
Stop Signal ◽  
Acute Administration ◽  
Double Blind ◽  
Outcome Valence ◽  
Lower Accuracy ◽  
Relevant Dose ◽  
Stop Signal Reaction Time

AbstractSerotonin is well known to affect the multifaceted construct of impulsivity. Lowering brain serotonin levels is shown to increase impulsive choice in delay-discounting tasks (1) but improves response inhibition in stop-signal paradigms. (2) Administration of the antidepressant citalopram in healthy people increases tendency to perform go choices in a Go/No-Go task independent of outcome valence (3). It is rather unclear thought how serotonergic neurotransmission affects several aspects of cognition. We administered a single dose of 20 mg escitalopram, a selective serotonin reuptake inhibitor, to 66 healthy participants, aged 18–45 years old, in a double-blind, randomized, placebo-controlled, parallel-groups study. Acute escitalopram administration had a beneficial effect on inhibitory control with reduced stop-signal reaction time observed in the treatment group. Participants made significantly more errors in a probabilistic learning task and had lower accuracy during the discrimination stage in an instrumental learning task thus indicating a learning impairment. More errors in the CANTAB intra-extra dimensional set shift task were also observed in the escitalopram-treated group. Our findings following acute administration of a clinically relevant dose of escitalopram show a dissociate role for serotonin in modulating cognition mediated by a potentially differential modulation of fronto-striatal loops.Disclosure of interestThe authors have not supplied their declaration of competing interest.

scholarly journals Mental model-based repeated multifaceted (MRM) intervention design: a conceptual framework for improving preventive health behaviors and outcomes

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Mazbahul G. Ahamad ◽  
Fahian Tanin
Keyword(s):  
Experimental Design ◽  
Health Education ◽  
Health Behaviors ◽  
Conceptual Framework ◽  
Mental Model ◽  
Preventive Health ◽  
Outcome Analysis ◽  
Intervention Design ◽  
Preventive Health Behaviors ◽  
Model Based

Abstract Objective Field interventions employed to improve preventive health behaviors and outcomes generally use well-established approaches; however, recent studies have reported that health education and promotional interventions have little to no impact on health behaviors, especially in low- and middle-income countries. We aimed to develop a conceptual framework to improve intervention designs that would internalize these concerns and limitations. Results We identified three major experimental design- and implementation-related concerns associated with mental models, including the balance between the treatment and control groups, the treatment group’s willingness to adopt suggested behaviors, and the type, length, frequency, intensity, and sequence of treatments. To minimize the influence of these aspects of an experimental design, we proposed a mental model-based repeated multifaceted (MRM) intervention design framework, which represents a supportive intervention design for the improvement of health education and promotional programs. The framework offers a step-by-step method that can be used for experimental and treatment design and outcome analysis, and that addresses potential implementation challenges.

Sign in / Sign up

Export Citation Format

Share Document