Electrophysiological signatures of hierarchical learning. Liu, Meng, Dong, W., Qin, S., Verguts, T., & Chen, Q. CEREBRAL CORTEX, 32(3):626–639, 2022. ![Electrophysiological signatures of hierarchical learning [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Human perception and learning is thought to rely on a hierarchical generative model that is continuously updated via precision-weighted prediction errors (pwPEs). However, the neural basis of such cognitive process and how it unfolds during decision-making remain poorly understood. To investigate this question, we combined a hierarchical Bayesian model (i.e., Hierarchical Gaussian Filter [HGF]) with electroencephalography (EEG), while participants performed a probabilistic reversal learning task in alternatingly stable and volatile environments. Behaviorally, the HGF fitted significantly better than two control, nonhierarchical, models. Neurally, low-level and high-level pwPEs were independently encoded by the P300 component. Low-level pwPEs were reflected in the theta (4-8 Hz) frequency band, but high-level pwPEs were not. Furthermore, the expressions of high-level pwPEs were stronger for participants with better HGF fit. These results indicate that the brain employs hierarchical learning and encodes both low- and high-level learning signals separately and adaptively.
@article{8728459,
abstract = {{Human perception and learning is thought to rely on a hierarchical generative model that is continuously updated via precision-weighted prediction errors (pwPEs). However, the neural basis of such cognitive process and how it unfolds during decision-making remain poorly understood. To investigate this question, we combined a hierarchical Bayesian model (i.e., Hierarchical Gaussian Filter [HGF]) with electroencephalography (EEG), while participants performed a probabilistic reversal learning task in alternatingly stable and volatile environments. Behaviorally, the HGF fitted significantly better than two control, nonhierarchical, models. Neurally, low-level and high-level pwPEs were independently encoded by the P300 component. Low-level pwPEs were reflected in the theta (4-8 Hz) frequency band, but high-level pwPEs were not. Furthermore, the expressions of high-level pwPEs were stronger for participants with better HGF fit. These results indicate that the brain employs hierarchical learning and encodes both low- and high-level learning signals separately and adaptively.}},
author = {{Liu, Meng and Dong, Wenshan and Qin, Shaozheng and Verguts, Tom and Chen, Qi}},
issn = {{1047-3211}},
journal = {{CEREBRAL CORTEX}},
keywords = {{Cognitive Neuroscience,computational modeling,EEG,hierarchical learning,precision-weighted prediction error,PREDICTION ERRORS,ANTERIOR CINGULATE,FRONTAL THETA,NEURAL MECHANISMS,BEHAVIOR,NEGATIVITY,MIDBRAIN,FEEDBACK,CORTEX,MODEL}},
language = {{eng}},
number = {{3}},
pages = {{626--639}},
title = {{Electrophysiological signatures of hierarchical learning}},
url = {{http://doi.org/10.1093/cercor/bhab245}},
volume = {{32}},
year = {{2022}},
}
Downloads: 0
{"_id":"BRGu3xounuYXr5fmH","bibbaseid":"liu-dong-qin-verguts-chen-electrophysiologicalsignaturesofhierarchicallearning-2022","author_short":["Liu, Meng","Dong, W.","Qin, S.","Verguts, T.","Chen, Q."],"bibdata":{"bibtype":"article","type":"article","abstract":"Human perception and learning is thought to rely on a hierarchical generative model that is continuously updated via precision-weighted prediction errors (pwPEs). However, the neural basis of such cognitive process and how it unfolds during decision-making remain poorly understood. To investigate this question, we combined a hierarchical Bayesian model (i.e., Hierarchical Gaussian Filter [HGF]) with electroencephalography (EEG), while participants performed a probabilistic reversal learning task in alternatingly stable and volatile environments. Behaviorally, the HGF fitted significantly better than two control, nonhierarchical, models. Neurally, low-level and high-level pwPEs were independently encoded by the P300 component. Low-level pwPEs were reflected in the theta (4-8 Hz) frequency band, but high-level pwPEs were not. Furthermore, the expressions of high-level pwPEs were stronger for participants with better HGF fit. These results indicate that the brain employs hierarchical learning and encodes both low- and high-level learning signals separately and adaptively.","author":[{"firstnames":[],"propositions":[],"lastnames":["Liu, Meng"],"suffixes":[]},{"propositions":[],"lastnames":["Dong"],"firstnames":["Wenshan"],"suffixes":[]},{"propositions":[],"lastnames":["Qin"],"firstnames":["Shaozheng"],"suffixes":[]},{"propositions":[],"lastnames":["Verguts"],"firstnames":["Tom"],"suffixes":[]},{"propositions":[],"lastnames":["Chen"],"firstnames":["Qi"],"suffixes":[]}],"issn":"1047-3211","journal":"CEREBRAL CORTEX","keywords":"Cognitive Neuroscience,computational modeling,EEG,hierarchical learning,precision-weighted prediction error,PREDICTION ERRORS,ANTERIOR CINGULATE,FRONTAL THETA,NEURAL MECHANISMS,BEHAVIOR,NEGATIVITY,MIDBRAIN,FEEDBACK,CORTEX,MODEL","language":"eng","number":"3","pages":"626–639","title":"Electrophysiological signatures of hierarchical learning","url":"http://doi.org/10.1093/cercor/bhab245","volume":"32","year":"2022","bibtex":"@article{8728459,\n abstract = {{Human perception and learning is thought to rely on a hierarchical generative model that is continuously updated via precision-weighted prediction errors (pwPEs). However, the neural basis of such cognitive process and how it unfolds during decision-making remain poorly understood. To investigate this question, we combined a hierarchical Bayesian model (i.e., Hierarchical Gaussian Filter [HGF]) with electroencephalography (EEG), while participants performed a probabilistic reversal learning task in alternatingly stable and volatile environments. Behaviorally, the HGF fitted significantly better than two control, nonhierarchical, models. Neurally, low-level and high-level pwPEs were independently encoded by the P300 component. Low-level pwPEs were reflected in the theta (4-8 Hz) frequency band, but high-level pwPEs were not. Furthermore, the expressions of high-level pwPEs were stronger for participants with better HGF fit. These results indicate that the brain employs hierarchical learning and encodes both low- and high-level learning signals separately and adaptively.}},\n author = {{Liu, Meng and Dong, Wenshan and Qin, Shaozheng and Verguts, Tom and Chen, Qi}},\n issn = {{1047-3211}},\n journal = {{CEREBRAL CORTEX}},\n keywords = {{Cognitive Neuroscience,computational modeling,EEG,hierarchical learning,precision-weighted prediction error,PREDICTION ERRORS,ANTERIOR CINGULATE,FRONTAL THETA,NEURAL MECHANISMS,BEHAVIOR,NEGATIVITY,MIDBRAIN,FEEDBACK,CORTEX,MODEL}},\n language = {{eng}},\n number = {{3}},\n pages = {{626--639}},\n title = {{Electrophysiological signatures of hierarchical learning}},\n url = {{http://doi.org/10.1093/cercor/bhab245}},\n volume = {{32}},\n year = {{2022}},\n}\n\n","author_short":["Liu, Meng","Dong, W.","Qin, S.","Verguts, T.","Chen, Q."],"key":"8728459","id":"8728459","bibbaseid":"liu-dong-qin-verguts-chen-electrophysiologicalsignaturesofhierarchicallearning-2022","role":"author","urls":{"Paper":"http://doi.org/10.1093/cercor/bhab245"},"keyword":["Cognitive Neuroscience","computational modeling","EEG","hierarchical learning","precision-weighted prediction error","PREDICTION ERRORS","ANTERIOR CINGULATE","FRONTAL THETA","NEURAL MECHANISMS","BEHAVIOR","NEGATIVITY","MIDBRAIN","FEEDBACK","CORTEX","MODEL"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://biblio.ugent.be/person/801001536697/publication/export?format=bibtex","dataSources":["TC7BryQp3aBdfCoH3"],"keywords":["cognitive neuroscience","computational modeling","eeg","hierarchical learning","precision-weighted prediction error","prediction errors","anterior cingulate","frontal theta","neural mechanisms","behavior","negativity","midbrain","feedback","cortex","model"],"search_terms":["electrophysiological","signatures","hierarchical","learning","liu","dong","qin","verguts","chen"],"title":"Electrophysiological signatures of hierarchical learning","year":2022}