{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T14:56:01Z","timestamp":1778165761666,"version":"3.51.4"},"reference-count":129,"publisher":"MIT Press","issue":"3","license":[{"start":{"date-parts":[[2021,7,9]],"date-time":"2021-07-09T00:00:00Z","timestamp":1625788800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2021,11,3]]},"abstract":"<jats:p>Word embeddings are vectorial semantic representations built with either counting or predicting techniques aimed at capturing shades of meaning from word co-occurrences. Since their introduction, these representations have been criticized for lacking interpretable dimensions. This property of word embeddings limits our understanding of the semantic features they actually encode. Moreover, it contributes to the \u201cblack box\u201d nature of the tasks in which they are used, since the reasons for word embedding performance often remain opaque to humans. In this contribution, we explore the semantic properties encoded in word embeddings by mapping them onto interpretable vectors, consisting of explicit and neurobiologically motivated semantic features (Binder et al. 2016). Our exploration takes into account different types of embeddings, including factorized count vectors and predict models (Skip-Gram, GloVe, etc.), as well as the most recent contextualized representations (i.e., ELMo and BERT).<\/jats:p><jats:p>In our analysis, we first evaluate the quality of the mapping in a retrieval task, then we shed light on the semantic features that are better encoded in each embedding type. A large number of probing tasks is finally set to assess how the original and the mapped embeddings perform in discriminating semantic categories. For each probing task, we identify the most relevant semantic features and we show that there is a correlation between the embedding performance and how they encode those features. This study sets itself as a step forward in understanding which aspects of meaning are captured by vector spaces, by proposing a new and simple method to carve human-interpretable semantic representations from distributional vectors.<\/jats:p>","DOI":"10.1162\/coli_a_00412","type":"journal-article","created":{"date-parts":[[2021,7,10]],"date-time":"2021-07-10T05:44:16Z","timestamp":1625895856000},"page":"663-698","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":35,"title":["Decoding Word Embeddings with Brain-Based Semantic Features"],"prefix":"10.1162","volume":"47","author":[{"given":"Emmanuele","family":"Chersoni","sequence":"first","affiliation":[{"name":"The Hong Kong Polytechnic University, Department of Chinese and Bilingual Studies. emmanuele.chersoni@polyu.edu.hk"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Enrico","family":"Santus","sequence":"additional","affiliation":[{"name":"MIT Computer Science and Artificial Intelligence Laboratory. esantus@mit.edu"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chu-Ren","family":"Huang","sequence":"additional","affiliation":[{"name":"The Hong Kong Polytechnic University, Department of Chinese and Bilingual Studies. churen.huang@polyu.edu.hk"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alessandro","family":"Lenci","sequence":"additional","affiliation":[{"name":"University of Pisa, Department of Philology, Literature and Linguistics. alessandro.lenci@unipi.it"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","published-online":{"date-parts":[[2021,11,3]]},"reference":[{"key":"2021111022501039600_bib1","doi-asserted-by":"crossref","first-page":"57","DOI":"10.18653\/v1\/W18-0107","article-title":"Experiential, distributional and dependency-based word embeddings have complementary roles in decoding brain activity","volume-title":"Proceedings of the 8th Workshop on Cognitive Modeling and Computational Linguistics (CMCL 2018)","author":"Abnar","year":"2018"},{"key":"2021111022501039600_bib2","first-page":"1","article-title":"Fine-grained analysis of sentence embeddings using auxiliary prediction tasks","volume-title":"Proceedings of ICLR","author":"Adi","year":"2017"},{"issue":"9","key":"2021111022501039600_bib3","doi-asserted-by":"publisher","first-page":"4379","DOI":"10.1093\/cercor\/bhw240","article-title":"Predicting neural activity patterns associated with sentences using a neurobiologically motivated model of semantic representation","volume":"27","author":"Anderson","year":"2016","journal-title":"Cerebral Cortex"},{"issue":"6","key":"2021111022501039600_bib4","doi-asserted-by":"publisher","first-page":"2396","DOI":"10.1093\/cercor\/bhy110","article-title":"Multiple regions of a cortical network commonly encode the meaning of words in multiple grammatical positions of read sentences","volume":"29","author":"Anderson","year":"2018","journal-title":"Cerebral Cortex"},{"key":"2021111022501039600_bib5","first-page":"2867","article-title":"Neural activation semantic models: Computational lexical semantic models of localized neural activations","volume-title":"Proceedings of COLING","author":"Athanasiou","year":"2018"},{"key":"2021111022501039600_bib6","first-page":"2200","article-title":"Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining","volume-title":"Proceedings of LREC","author":"Baccianella","year":"2010"},{"key":"2021111022501039600_bib7","article-title":"Can eye movement data be used as ground truth for word embeddings evaluation?","volume-title":"Proceedings of the LREC Workshop on Linguistic and Neurocognitive Resources","author":"Bakarov","year":"2018"},{"issue":"3","key":"2021111022501039600_bib8","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1007\/s10579-009-9081-4","article-title":"The WaCky Wide Web: A collection of very large linguistically processed web-crawled corpora","volume":"43","author":"Baroni","year":"2009","journal-title":"Language Resources and Evaluation"},{"key":"2021111022501039600_bib9","first-page":"238","article-title":"Don\u2019t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors","volume-title":"Proceedings of ACL","author":"Baroni","year":"2014"},{"issue":"4","key":"2021111022501039600_bib10","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1162\/coli_a_00016","article-title":"Distributional memory: A general framework for corpus-based semantics","volume":"36","author":"Baroni","year":"2010","journal-title":"Computational Linguistics"},{"key":"2021111022501039600_bib11","article-title":"Robust evaluation of language-brain encoding experiments","author":"Beinborn","year":"2019","journal-title":"arXiv preprint arXiv:1904.02547"},{"issue":"3-4","key":"2021111022501039600_bib12","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1080\/02643294.2016.1147426","article-title":"Toward a brain-based componential semantic representation","volume":"33","author":"Binder","year":"2016","journal-title":"Cognitive Neuropsychology"},{"key":"2021111022501039600_bib13","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1162\/tacl_a_00051","article-title":"Enriching word vectors with subword information","volume":"5","author":"Bojanowski","year":"2017","journal-title":"Transactions of the ACL"},{"key":"2021111022501039600_bib14","doi-asserted-by":"publisher","first-page":"213","DOI":"10.1146\/annurev-linguistics-011619-030303","article-title":"Distributional semantics and linguistic theory","volume":"6","author":"Boleda","year":"2020","journal-title":"Annual Review of Linguistics"},{"key":"2021111022501039600_bib15","first-page":"2","article-title":"Distributional semantic features as semantic primitives - or not","volume-title":"Proceedings of Knowledge Representation and Reasoning: Integrating Symbolic and Neural Approaches: Papers from the 2015 AAAI Spring Symposium","author":"Boleda","year":"2015"},{"key":"2021111022501039600_bib16","first-page":"4758","article-title":"Interpreting pretrained contextualized representations via reductions to static embeddings","volume-title":"Proceedings of ACL","author":"Bommasani","year":"2020"},{"key":"2021111022501039600_bib17","article-title":"Affective Norms for English Words (ANEW)","volume-title":"Technical Report C-3. UF Center for the Study of Emotion and Attention","author":"Bradley","year":"2017"},{"key":"2021111022501039600_bib18","first-page":"2892","article-title":"Emotion representation mapping for automatic lexicon construction (mostly) performs on human level","volume-title":"Proceedings of COLING","author":"Buechel","year":"2018"},{"key":"2021111022501039600_bib19","first-page":"523","article-title":"Modelling metaphor with attribute-based semantics","volume-title":"Proceedings of EACL","author":"Bulat","year":"2017"},{"key":"2021111022501039600_bib20","first-page":"1081","article-title":"Speaking, seeing, understanding: Correlating semantic models with conceptual representation in the brain","volume-title":"Proceedings of EMNLP","author":"Bulat","year":"2017"},{"key":"2021111022501039600_bib21","first-page":"579","article-title":"Vision and feature norms: Improving automatic feature norm learning through cross-modal maps","volume-title":"Proceedings of NAACL-HLT","author":"Bulat","year":"2016"},{"issue":"3","key":"2021111022501039600_bib22","doi-asserted-by":"publisher","first-page":"890","DOI":"10.3758\/s13428-011-0183-8","article-title":"Extracting semantic representations from word co-occurrence statistics: Stop-lists, stemming, and SVD","volume":"44","author":"Bullinaria","year":"2012","journal-title":"Behavior Research Methods"},{"key":"2021111022501039600_bib23","doi-asserted-by":"crossref","first-page":"37","DOI":"10.18653\/v1\/W16-0409","article-title":"Sentiment lexicon creation using continuous latent space and neural networks","volume-title":"Proceedings of the NAACL Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis","author":"Cardoso","year":"2016"},{"issue":"1","key":"2021111022501039600_bib24","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1093\/cercor\/bhw379","article-title":"Representational similarity mapping of distributional semantics in left inferior frontal, middle temporal, and motor cortex","volume":"27","author":"Carota","year":"2017","journal-title":"Cerebral Cortex"},{"issue":"2","key":"2021111022501039600_bib25","doi-asserted-by":"publisher","first-page":"716","DOI":"10.1016\/j.neuroimage.2010.04.271","article-title":"Quantitative modeling of the neural representation of objects: How semantic feature norms can account for fMRI activation","volume":"56","author":"Chang","year":"2011","journal-title":"NeuroImage"},{"key":"2021111022501039600_bib26","article-title":"One billion word benchmark for measuring progress in statistical language modeling","author":"Chelba","year":"2013","journal-title":"arXiv preprint arXiv:1312.3005"},{"key":"2021111022501039600_bib27","first-page":"5708","article-title":"Are word embeddings really a bad fit for the estimation of thematic fit?","volume-title":"Proceedings of LREC","author":"Chersoni","year":"2020"},{"issue":"4","key":"2021111022501039600_bib28","doi-asserted-by":"publisher","first-page":"483","DOI":"10.1017\/S1351324919000214","article-title":"A structured distributional model of sentence meaning and processing","volume":"25","author":"Chersoni","year":"2019","journal-title":"Natural Language Engineering"},{"key":"2021111022501039600_bib29","first-page":"227","article-title":"When is a bishop not like a rook? When it\u2019s like a rabbi! Multi-prototype BERT embeddings for estimating semantic relationships","volume-title":"Proceedings of CoNLL 2020","author":"Chronis","year":"2020"},{"key":"2021111022501039600_bib30","first-page":"2126","article-title":"What you can cram into a single $&!#* vector: Probing sentence embeddings for linguistic properties","volume-title":"Proceedings of ACL","author":"Conneau","year":"2018"},{"key":"2021111022501039600_bib31","first-page":"1","article-title":"Not all moods are created equal! Exploring human emotional states in social media","volume-title":"Proceedings of ICWSM","author":"De Choudhury","year":"2012"},{"key":"2021111022501039600_bib32","first-page":"5853","article-title":"Feature2Vec: Distributional semantic modelling of human property knowledge","volume-title":"Proceedings of EMNLP","author":"Derby","year":"2019"},{"key":"2021111022501039600_bib33","first-page":"70","article-title":"Using fMRI activation to conceptual stimuli to evaluate methods for extracting conceptual representations from corpora","volume-title":"Proceedings of the NAACL Workshop on Computational Neurolinguistics","author":"Devereux","year":"2010"},{"issue":"4","key":"2021111022501039600_bib34","doi-asserted-by":"publisher","first-page":"1119","DOI":"10.3758\/s13428-013-0420-4","article-title":"The Centre for Speech, Language and the Brain (CSLB) concept property norms","volume":"46","author":"Devereux","year":"2014","journal-title":"Behavior Research Methods"},{"key":"2021111022501039600_bib35","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"Proceedings of NAACL-HLT 2019","author":"Devlin","year":"2019"},{"key":"2021111022501039600_bib36","first-page":"5155","article-title":"Modeling affirmative and negated action processing in the brain with lexical and compositional semantic models","volume-title":"Proceedings of ACL","author":"Djokic","year":"2019"},{"issue":"4","key":"2021111022501039600_bib37","doi-asserted-by":"publisher","first-page":"723","DOI":"10.1162\/coli_a_00017","article-title":"A flexible, corpus-driven model of regular and inverse selectional preferences","volume":"36","author":"Erk","year":"2010","journal-title":"Computational Linguistics"},{"key":"2021111022501039600_bib38","first-page":"417","article-title":"Sentiwordnet: A publicly available lexical resource for opinion mining","volume-title":"Proceedings of LREC","author":"Esuli","year":"2006"},{"key":"2021111022501039600_bib39","doi-asserted-by":"crossref","first-page":"134","DOI":"10.18653\/v1\/W16-2524","article-title":"Probing for semantic evidence of composition by means of simple classification tasks","volume-title":"Proceedings of the 1st Workshop on Evaluating Vector Space Representations for NLP","author":"Ettinger","year":"2016"},{"key":"2021111022501039600_bib40","first-page":"52","article-title":"From distributional semantics to feature norms: Grounding semantic models in human perceptual data","volume-title":"Proceedings of IWCS","author":"F\u0103g\u0103r\u0103san","year":"2015"},{"key":"2021111022501039600_bib41","article-title":"Does the brain represent words? An evaluation of brain decoding studies of language understanding","author":"Gauthier","year":"2018","journal-title":"arXiv preprint arXiv:1806.00591"},{"key":"2021111022501039600_bib42","article-title":"Evaluating semantic models with word-sentence relatedness","author":"Glasgow","year":"2016","journal-title":"arXiv preprint arXiv:1603.07253"},{"key":"2021111022501039600_bib43","article-title":"Semantic vector space models predict neural responses to complex visual stimuli","author":"G\u00fc\u00e7l\u00fc","year":"2015","journal-title":"arXiv preprint arXiv:1510.04738"},{"key":"2021111022501039600_bib44","first-page":"4129","article-title":"A structural probe for finding syntax in word representations","volume-title":"Proceedings of NAACL","author":"Hewitt","year":"2019"},{"issue":"4","key":"2021111022501039600_bib45","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1162\/COLI_a_00237","article-title":"SimLex-999: Evaluating semantic models with (genuine) similarity estimation","volume":"41","author":"Hill","year":"2015","journal-title":"Computational Linguistics"},{"key":"2021111022501039600_bib46","first-page":"77","article-title":"Distributed representations","volume-title":"Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Volume 1: Foundations","author":"Hinton","year":"1986"},{"key":"2021111022501039600_bib47","first-page":"538","article-title":"CogniVal: A framework for cognitive word embedding evaluation","volume-title":"Proceedings of CONLL","author":"Hollenstein","year":"2019"},{"key":"2021111022501039600_bib48","first-page":"328","article-title":"Universal language model fine-tuning for text classification","volume-title":"Proceedings of ACL","author":"Howard","year":"2018"},{"issue":"7600","key":"2021111022501039600_bib49","doi-asserted-by":"publisher","first-page":"453","DOI":"10.1038\/nature17637","article-title":"Natural speech reveals the semantic maps that tile human cerebral cortex","volume":"532","author":"Huth","year":"2016","journal-title":"Nature"},{"key":"2021111022501039600_bib50","volume-title":"Semantic Structures","author":"Jackendoff","year":"1990"},{"key":"2021111022501039600_bib51","first-page":"3651","article-title":"What does BERT learn about the structure of language?","volume-title":"Proceedings of ACL","author":"Jawahar","year":"2019"},{"key":"2021111022501039600_bib52","first-page":"52","article-title":"Verb argument structure alternations in word and sentence embeddings","volume-title":"Proceedings of SCIL","author":"Kann","year":"2019"},{"key":"2021111022501039600_bib53","first-page":"235","article-title":"Probing what different NLP tasks teach machines about function word comprehension","volume-title":"Proceedings of *SEM","author":"Kim","year":"2019"},{"key":"2021111022501039600_bib54","first-page":"345","article-title":"Leveraging distributed representations and lexico-syntactic fixedness for token- level prediction of the idiomaticity of English verb-noun combinations","volume-title":"Proceedings of ACL","author":"King","year":"2018"},{"issue":"1","key":"2021111022501039600_bib55","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1007\/s10579-007-9048-2","article-title":"A large-scale classification of English verbs","volume":"42","author":"Kipper","year":"2008","journal-title":"Language Resource and Evaluation"},{"key":"2021111022501039600_bib56","first-page":"4801","article-title":"Spying on your neighbors: Fine-grained probing of contextual embeddings for information about surrounding words","volume-title":"Proceedings of ACL","author":"Klafka","year":"2020"},{"issue":"4","key":"2021111022501039600_bib57","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1017\/S1351324910000124","article-title":"Directional distributional similarity for lexical inference","volume":"16","author":"Kotlerman","year":"2010","journal-title":"Journal of Natural Language Engineering"},{"issue":"2","key":"2021111022501039600_bib58","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1037\/0033-295X.104.2.211","article-title":"A solution to Plato\u2019s problem: The latent semantic analysis theory of acquisition, induction, and representation of knowledge","volume":"104","author":"Landauer","year":"1997","journal-title":"Psychological Review"},{"key":"2021111022501039600_bib59","doi-asserted-by":"crossref","DOI":"10.4324\/9780203936399","volume-title":"Handbook of Latent Semantic Analysis","author":"Landauer","year":"2007"},{"key":"2021111022501039600_bib60","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1111\/tops.12335","article-title":"Composing and updating verb argument expectations: A distributional semantic model","volume-title":"Proceedings of ACL Workshop on Cognitive Modeling and Computational Linguistics","author":"Lenci","year":"2011"},{"key":"2021111022501039600_bib61","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1146\/annurev-linguistics-030514-125254","article-title":"Distributional models of word meaning","volume":"4","author":"Lenci","year":"2018","journal-title":"Annual Review of Linguistics"},{"issue":"3","key":"2021111022501039600_bib62","doi-asserted-by":"crossref","first-page":"550","DOI":"10.1111\/tops.12335","article-title":"The emotions of abstract words: A distributional semantic analysis","volume":"10","author":"Lenci","year":"2018","journal-title":"Topics in Cognitive Science"},{"key":"2021111022501039600_bib63","volume-title":"English Verb Classes and Alternations: A Preliminary Investigation","author":"Levin","year":"1993"},{"key":"2021111022501039600_bib64","first-page":"302","article-title":"Dependency-based word embeddings","volume-title":"Proceedings of ACL","author":"Levy","year":"2014"},{"key":"2021111022501039600_bib65","first-page":"211","article-title":"Improving distributional similarity with lessons learned from word embeddings","volume-title":"Transactions of the ACL","author":"Levy","year":"2015"},{"key":"2021111022501039600_bib66","article-title":"Introduction","volume-title":"Proceedings of EMNLP Workshop on BlackBoxNLP: Analyzing and Interpreting Neural Networks for NLP","author":"Linzen","year":"2018"},{"key":"2021111022501039600_bib67","article-title":"Introduction","volume-title":"Proceedings of ACL Workshop on BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","author":"Linzen","year":"2019"},{"key":"2021111022501039600_bib68","first-page":"1073","article-title":"Linguistic knowledge and transferability of contextual representations","volume-title":"Proceedings of NAACL","author":"Liu","year":"2019"},{"issue":"4","key":"2021111022501039600_bib69","doi-asserted-by":"publisher","first-page":"838","DOI":"10.3758\/PBR.15.4.838","article-title":"Embodied relations are encoded in language","volume":"15","author":"Louwerse","year":"2008","journal-title":"Psychonomic Bulletin & Review"},{"key":"2021111022501039600_bib70","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.jml.2016.04.001","article-title":"Explaining human performance in psycholinguistic tasks with models of semantic similarity based on prediction and counting: A review and empirical validation","volume":"92","author":"Mandera","year":"2017","journal-title":"Journal of Memory and Language"},{"key":"2021111022501039600_bib71","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511809071","volume-title":"Introduction to Information Retrieval","author":"Manning","year":"2008"},{"key":"2021111022501039600_bib72","first-page":"55","article-title":"The Stanford CoreNLP natural language processing toolkit","volume-title":"Association for Computational Linguistics (ACL) System Demonstrations","author":"Manning","year":"2014"},{"key":"2021111022501039600_bib73","first-page":"6294","article-title":"Learned in translation: Contextualized word vectors","volume-title":"Advances in Neural Information Processing Systems","author":"McCann","year":"2017"},{"issue":"4","key":"2021111022501039600_bib74","doi-asserted-by":"publisher","first-page":"547","DOI":"10.3758\/BF03192726","article-title":"Semantic feature production norms for a large set of living and nonliving things","volume":"37","author":"McRae","year":"2005","journal-title":"Behavior Research Methods"},{"issue":"6","key":"2021111022501039600_bib75","doi-asserted-by":"publisher","first-page":"1417","DOI":"10.1111\/j.1749-818X.2009.00174.x","article-title":"People use their knowledge of common events to understand language, and do so as quickly as possible","volume":"3","author":"McRae","year":"2009","journal-title":"Language and Linguistics Compass"},{"key":"2021111022501039600_bib76","article-title":"Efficient estimation of word representations in vector space","volume-title":"Proceedings of ICLR","author":"Mikolov","year":"2013"},{"issue":"5880","key":"2021111022501039600_bib77","doi-asserted-by":"publisher","first-page":"1191","DOI":"10.1126\/science.1152876","article-title":"Predicting human brain activity associated with the meanings of nouns","volume":"320","author":"Mitchell","year":"2008","journal-title":"Science"},{"key":"2021111022501039600_bib78","first-page":"114","article-title":"Selecting corpus-semantic models for neurolinguistic decoding","volume-title":"Proceedings of *SEM","author":"Murphy","year":"2012"},{"key":"2021111022501039600_bib79","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/1602.001.0001","volume-title":"The Big Book of Concepts","author":"Murphy","year":"2002"},{"key":"2021111022501039600_bib80","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511780684","volume-title":"Lexical Meaning","author":"Murphy","year":"2010"},{"issue":"2","key":"2021111022501039600_bib81","doi-asserted-by":"publisher","first-page":"400","DOI":"10.1016\/j.neuroimage.2010.07.073","article-title":"Encoding and decoding in fMRI","volume":"56","author":"Naselaris","year":"2011","journal-title":"NeuroImage"},{"key":"2021111022501039600_bib82","article-title":"A new ANEW: Evaluation of a word list for sentiment analysis in microblogs","author":"Nielsen","year":"2011","journal-title":"arXiv preprint arXiv:1103.2903"},{"key":"2021111022501039600_bib83","first-page":"315","article-title":"VerbNet: Capturing English verb behavior, meaning and usage","author":"Palmer","year":"2017","journal-title":"The Oxford Handbook of Cognitive Science"},{"key":"2021111022501039600_bib84","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2021.starsem-1.1","article-title":"Did the cat drink the coffee? Challenging transformers with generalized event knowledge","volume-title":"Proceedings of *SEM","author":"Pedinotti","year":"2021"},{"key":"2021111022501039600_bib85","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"2021111022501039600_bib86","first-page":"1532","article-title":"GloVe: Global vectors for word representation","volume-title":"Proceedings of EMNLP","author":"Pennington","year":"2014"},{"key":"2021111022501039600_bib87","doi-asserted-by":"publisher","first-page":"240","DOI":"10.1038\/s41467-018-03068-4","article-title":"Using Wikipedia to learn semantic feature representations of concrete concepts in neuroimaging experiments","volume":"194","author":"Pereira","year":"2013","journal-title":"Artificial Intelligence"},{"key":"2021111022501039600_bib88","doi-asserted-by":"crossref","first-page":"72","DOI":"10.3389\/fnhum.2011.00072","article-title":"Generating text from functional brain images","volume":"5","author":"Pereira","year":"2011","journal-title":"Frontiers in Human Neuroscience"},{"issue":"1","key":"2021111022501039600_bib89","doi-asserted-by":"publisher","first-page":"963","DOI":"10.1016\/j.neuron.2011.11.001","article-title":"Toward a universal decoder of linguistic meaning from brain activation","volume":"9","author":"Pereira","year":"2018","journal-title":"Nature Communications"},{"key":"2021111022501039600_bib90","first-page":"2227","article-title":"Deep contextualized word representations","volume-title":"Proceedings of NAACL-HLT","author":"Peters","year":"2018"},{"issue":"5","key":"2021111022501039600_bib91","doi-asserted-by":"publisher","first-page":"692","DOI":"10.1016\/j.neuron.2011.11.001","article-title":"Inferring mental states from neuroimaging data: from reverse inference to large-scale decoding","volume":"72","author":"Poldrack","year":"2011","journal-title":"Neuron"},{"key":"2021111022501039600_bib92","doi-asserted-by":"crossref","DOI":"10.1017\/9780511982378","volume-title":"The Lexicon","author":"Pustejovsky","year":"2019"},{"issue":"8","key":"2021111022501039600_bib93","doi-asserted-by":"publisher","first-page":"1584","DOI":"10.1080\/17470218.2014.941296","article-title":"Reproducing affective norms with lexical co-occurrence statistics: Predicting valence, arousal, and dominance","volume":"68","author":"Recchia","year":"2015","journal-title":"The Quarterly Journal of Experimental Psychology"},{"issue":"2","key":"2021111022501039600_bib94","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1111\/j.1756-8765.2010.01111.x","article-title":"Redundancy in perceptual and linguistic experience: Comparing feature-based and distributional models of semantic representation","volume":"3","author":"Riordan","year":"2011","journal-title":"Topics in Cognitive Science"},{"key":"2021111022501039600_bib95","first-page":"2890","article-title":"Verbal multiword expressions for identification of metaphor","volume-title":"Proceedings of ACL","author":"Rohanian","year":"2020"},{"key":"2021111022501039600_bib96","first-page":"33","article-title":"The distributional hypothesis","volume":"20","author":"Sahlgren","year":"2008","journal-title":"Italian Journal of Linguistics"},{"key":"2021111022501039600_bib97","first-page":"648","article-title":"Measuring thematic fit with distributional feature overlap","volume-title":"Proceedings of EMNLP","author":"Santus","year":"2017"},{"key":"2021111022501039600_bib98","doi-asserted-by":"crossref","first-page":"99","DOI":"10.18653\/v1\/W16-2518","article-title":"Thematic fit evaluation: An aspect of selectional preferences","volume-title":"Proceedings of the ACL Workshop on Evaluating Vector-Space Representations for NLP","author":"Sayeed","year":"2016"},{"key":"2021111022501039600_bib99","first-page":"43","article-title":"Understanding language-elicited EEG data by predicting it from a fine-tuned language model","volume-title":"Proceedings of NAACL","author":"Schwartz","year":"2019"},{"key":"2021111022501039600_bib100","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18653\/v1\/W19-2001","article-title":"Neural vector conceptualization for word vector space interpretation","volume-title":"Proceedings of the NAACL Workshop on Evaluating Vector Space Representations","author":"Schwarzenberg","year":"2019"},{"key":"2021111022501039600_bib101","first-page":"346","article-title":"Automatic domain adaptation outperforms manual domain adaptation for predicting financial outcomes","volume-title":"Proceedings of ACL","author":"Sedinkina","year":"2019"},{"issue":"10","key":"2021111022501039600_bib102","doi-asserted-by":"publisher","first-page":"1769","DOI":"10.1109\/TASLP.2018.2837384","article-title":"Semantic structure and interpretability of word embeddings","volume":"26","author":"\u015eenel","year":"2018","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"2021111022501039600_bib103","doi-asserted-by":"publisher","first-page":"403","DOI":"10.1162\/tacl_a_00277","article-title":"Still a pain in the neck: Evaluating text representations on lexical composition","volume":"7","author":"Shwartz","year":"2019","journal-title":"Transactions of the ACL"},{"key":"2021111022501039600_bib104","first-page":"295","article-title":"Frame identification as categorization: exemplars vs prototypes in embeddingland","volume-title":"Proceedings of IWCS","author":"Sikos","year":"2019"},{"key":"2021111022501039600_bib105","doi-asserted-by":"crossref","first-page":"116","DOI":"10.18653\/v1\/W16-2521","article-title":"Evaluating word embeddings with fMRI and eye-tracking","volume-title":"Proceedings of the ACL Workshop on Evaluating Vector-Space Representations for NLP","author":"S\u00f8gaard","year":"2016"},{"key":"2021111022501039600_bib106","first-page":"7047","article-title":"Towards sentence-level brain decoding with distributed representations","volume-title":"Proceedings of AAAI","author":"Sun","year":"2019"},{"key":"2021111022501039600_bib107","first-page":"1511","article-title":"Sensicon: An automatically constructed sensorial lexicon","volume-title":"Proceedings of EMNLP","author":"Tekiroglu","year":"2014"},{"key":"2021111022501039600_bib108","first-page":"235","article-title":"What do you learn from context? Probing for sentence structure in contextualized word representations","volume-title":"Proceedings of ICLR 2019","author":"Tenney","year":"2019"},{"issue":"4","key":"2021111022501039600_bib109","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1145\/944012.944013","article-title":"Measuring praise and criticism: Inference of semantic orientation from association","volume":"21","author":"Turney","year":"2003","journal-title":"ACM Transactions on Information Systems (TOIS)"},{"key":"2021111022501039600_bib110","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1613\/jair.2934","article-title":"From frequency to meaning: Vector space models of semantics","volume":"37","author":"Turney","year":"2010","journal-title":"Journal of Artificial Intelligence Research"},{"key":"2021111022501039600_bib111","first-page":"1","article-title":"Extrapolating Binder style word embeddings to new words","volume-title":"Proceedings of the LREC Workshop on Linguistic and Neurocognitive Resources","author":"Turton","year":"2020"},{"key":"2021111022501039600_bib112","first-page":"1145","article-title":"A neurobiologically motivated analysis of distributional semantic models","volume-title":"Proceedings of CogSci","author":"Utsumi","year":"2018"},{"issue":"6","key":"2021111022501039600_bib113","doi-asserted-by":"crossref","first-page":"e12844","DOI":"10.1111\/cogs.12844","article-title":"Exploring what is encoded in distributional word vectors: A neurobiologically motivated analysis","volume":"44","author":"Utsumi","year":"2020","journal-title":"Cognitive Science"},{"key":"2021111022501039600_bib114","first-page":"5998","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems","author":"Vaswani","year":"2017"},{"issue":"2","key":"2021111022501039600_bib115","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1515\/LANGCOG.2009.011","article-title":"Toward a theory of semantic representation","volume":"1","author":"Vigliocco","year":"2009","journal-title":"Language and Cognition"},{"key":"2021111022501039600_bib116","first-page":"195","article-title":"Semantic representation","volume-title":"The Oxford Handbook of Psycholinguistics","author":"Vigliocco","year":"2007"},{"issue":"1","key":"2021111022501039600_bib117","doi-asserted-by":"publisher","first-page":"183","DOI":"10.3758\/BRM.40.1.183","article-title":"Semantic feature production norms for a large set of objects and events","volume":"40","author":"Vinson","year":"2008","journal-title":"Behavior Research Methods"},{"issue":"4","key":"2021111022501039600_bib118","doi-asserted-by":"publisher","first-page":"781","DOI":"10.1162\/COLI_a_00301","article-title":"HyperLex: A large-scale evaluation of graded lexical entailment","volume":"43","author":"Vuli\u0107","year":"2017","journal-title":"Computational Linguistics"},{"key":"2021111022501039600_bib119","first-page":"7222","article-title":"Probing pretrained language models for lexical semantics","volume-title":"Proceedings of EMNLP","author":"Vuli\u0107","year":"2020"},{"key":"2021111022501039600_bib120","doi-asserted-by":"publisher","first-page":"625","DOI":"10.1162\/tacl_a_00290","article-title":"Neural network acceptability judgments","volume":"7","author":"Warstadt","year":"2019","journal-title":"Transactions of the ACL"},{"issue":"11","key":"2021111022501039600_bib121","doi-asserted-by":"crossref","first-page":"e112575","DOI":"10.1371\/journal.pone.0112575","article-title":"Simultaneously uncovering the patterns of brain regions involved in different story reading subprocesses","volume":"9","author":"Wehbe","year":"2014","journal-title":"PloS ONE"},{"key":"2021111022501039600_bib122","article-title":"Does BERT make any sense? Interpretable word sense disambiguation with contextualized embeddings","volume-title":"Proceedings of KONVENS","author":"Wiedemann","year":"2019"},{"key":"2021111022501039600_bib123","doi-asserted-by":"crossref","DOI":"10.1093\/oso\/9780198700029.001.0001","volume-title":"Semantics: Primes and Universals","author":"Wierzbicka","year":"1996"},{"key":"2021111022501039600_bib124","first-page":"5740","article-title":"Probing for semantic classes: Diagnosing the meaning content of word embeddings","volume-title":"Proceedings of ACL","author":"Yaghoobzadeh","year":"2019"},{"key":"2021111022501039600_bib125","first-page":"5753","article-title":"XLNet: Generalized autoregressive pretraining for language understanding","volume-title":"Advances in Neural Information Processing Systems 32","author":"Yang","year":"2019"},{"issue":"4","key":"2021111022501039600_bib126","doi-asserted-by":"publisher","first-page":"1015","DOI":"10.3758\/s13423-015-0948-7","article-title":"Putting concepts into context","volume":"23","author":"Yee","year":"2016","journal-title":"Psychonomic Bulletin & Review"},{"key":"2021111022501039600_bib127","first-page":"5247","article-title":"Multiplex word embeddings for selectional preference acquisition","volume-title":"Proceedings of EMNLP","author":"Zhang","year":"2019"},{"key":"2021111022501039600_bib128","first-page":"722","article-title":"SP-10K: A large-scale evaluation set for selectional preference acquisition","volume-title":"Proceedings of ACL","author":"Zhang","year":"2019"},{"key":"2021111022501039600_bib129","first-page":"19","article-title":"Aligning books and movies: Towards story-like visual explanations by watching movies and reading books","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Zhu","year":"2015"}],"container-title":["Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/47\/3\/663\/1971848\/coli_a_00412.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/coli\/article-pdf\/47\/3\/663\/1971848\/coli_a_00412.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,5]],"date-time":"2023-11-05T21:54:35Z","timestamp":1699221275000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/coli\/article\/47\/3\/663\/102823\/Decoding-Word-Embeddings-with-Brain-Based-Semantic"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11]]},"references-count":129,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2021,11,3]]},"published-print":{"date-parts":[[2021,11,3]]}},"URL":"https:\/\/doi.org\/10.1162\/coli_a_00412","relation":{},"ISSN":["0891-2017","1530-9312"],"issn-type":[{"value":"0891-2017","type":"print"},{"value":"1530-9312","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,11]]},"published":{"date-parts":[[2021,11]]}}}