Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Probing the contents of semantic representations from text, behavior, and brain data using the psychNorms metabase

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read Behavior-derived word vectors capture psychological content that text embeddings miss, complementing them on affective, agentic, and socio-moral dimensions.

desk verdict A useful norm metabase and the broadest representation comparison I've seen, but the unique-variance claim is likely inflated by selecting the top representations on the same norms used for evaluation. read the letter →

arxiv 2412.04936 v1 pith:CBG7VD43 submitted 2024-12-06 cs.CL cs.AIcs.LG

classification cs.CLcs.AIcs.LG
keywords semanticrepresentationsrepresentationalsimilarityanalysiscontentwordnormsfreeassociationsembeddingprobinghuman-alignedlanguagemodelspsych
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether word representations built from human behavior, such as free associations, and from brain recordings carry meaning that ordinary text-based embeddings miss. To answer it, the authors assembled a metabase of 292 human-rated word properties (psychNorms), introduced a probing framework called representational content analysis, and compared ten text, ten behavior, and six brain representations. They find that behavior-derived vectors, particularly PPMI SVD SWOW trained on the Small World of Words free-association data, rival text on many psychological dimensions and explain additional variance on affective, agency, and social/moral norms when added to text in an ensemble. The paper concludes that behavior is a practical complement to text for applications that aim to model human semantic representations, from sentiment analysis to evaluating large language models.

What carries the argument

The machinery that carries the argument is representational content analysis (RCA): L2-regularized linear probes are fit to predict each of 292 norms from each word-vector representation, yielding a psychological content profile per representation. The decisive test is the ensemble RCA, which concatenates the top text and top behavior representations and compares the marginal increase in cross-validated R2 against a text-plus-text ensemble, with all vocabularies subset to a common intersection so training-set size is matched. The behavior representation of interest, PPMI SVD SWOW, is built by applying positive pointwise mutual information and singular value decomposition to the Small World of Words cue–response matrix.

What would settle it

A concrete falsifying observation: if the Text & Behavior advantage over Text & Text disappears when the top behavior representation is selected using only a training subset of the 292 norms (or when both ensembles are matched for feature dimensionality, or when norms are collected from a non-verbal/implicit task), then the claim that behavior captures unique psychological variance would be undermined. Alternatively, training PPMI SVD on a random reshuffling of the SWOW association matrix while preserving marginal frequencies, and showing it yields a similar R2 gain in the ensemble, would indicate the gain is not due to semantic content.

Watch

Extended reading notes

Core claim

The central claim is that free-association-derived word vectors encode psychological information that text embeddings do not, and that this information is not redundant: ensembling the top text representation (CBOW GoogleNews or fastText CommonCrawl) with the top behavior representation (PPMI SVD SWOW) beats ensembling two text representations on affective norms (Dominance, Arousal, Valence, Emotion), agency norms (Goals/Needs, Motor), and social/moral norms, with median differences in R2 between 0.03 and 0.08, all significant at p < .05. The paper also reports representational similarity analysis showing clear clustering by data type rather than by learning algorithm, with behavior the most distinct of the three types. On this basis, the authors maintain that behavior representations, trained on orders of magnitude less data than text, are an important complement for measuring and modeling human representations.

Load-bearing premise

The load-bearing assumption is that the extra variance explained by adding the free-association vectors to text vectors reflects genuinely unique semantic content, rather than artifacts of choosing the best representations on the evaluation norms, having different feature dimensionality between the compared ensembles, or shared method variance between human ratings and human-elicited associations.

Editorial extensions

If this is right

  • Behavior-based semantic representations can supplement text embeddings in sentiment analysis, cognitive modeling, and other applications that depend on affective, agentic, or social-moral content.
  • The psychNorms metabase of 292 human-rated norms is a reusable resource for probing any word-level representation along psychologically meaningful dimensions.
  • Large language models trained or fine-tuned on structured behavioral data, such as free associations, could achieve better alignment with humans on psychological dimensions that text alone underrepresents.
  • Representational content analysis gives a general recipe for turning opaque word vectors into interpretable profiles, which can clarify what different models capture and where they diverge.
  • Expanding behavior-data collection efforts (e.g., more free associations) could bring behavior representations closer to text in coverage and performance, potentially improving their practical usefulness.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The unique-variance finding could be tested more stringently by selecting the top representations on a training split of the norms and evaluating on a held-out split; the current analysis selects representations using the same 292 norms it then evaluates on, which may inflate the reported gains.
  • Because behavior vectors and norm ratings both come from the same kind of human subjects, some of the shared variance may reflect common response styles rather than semantic content; collecting norms through implicit or task-based measures would clarify whether behavior truly adds semantic information.
  • The brain representations' poor performance in this study may be a byproduct of small vocabularies and crude word-level extraction from sentence recordings rather than a property of brain data generally; better word-level brain representations might change the comparisons.
  • RCA applied to non-English norms could reveal whether the text–behavior complementarity extends across languages or is specific to English datasets.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This paper presents a systematic comparison of semantic representations derived from text, behavior, and brain data, using representational similarity analysis (RSA) and a new interpretability framework called representational content analysis (RCA). The authors introduce psychNorms, a metabase of 292 word norms grouped into 27 categories, and probe 10 text, 10 behavior, and 6 brain representations against these norms. They report that behavior representations, particularly a PPMI-SVD representation trained on free associations, capture unique variance on affective, agentic, and socio-moral dimensions beyond text representations when combined in an ensemble. The paper also finds that brain representations contain little probe-able psychological content at the word level. The central claim is that behavior data complement text data for human-aligned semantic modeling.

Significance. If the central claim holds, the paper would provide the first large-scale evidence that behavior-derived word representations (e.g., from free associations) can complement text embeddings on psychologically meaningful dimensions, with implications for sentiment analysis, cognitive modeling, and LLM alignment. The psychNorms metabase and the RCA framework are valuable contributions, and the authors make code and data available. The nested cross-validation for probing is a technical strength. However, the selection-on-evaluation issue in Section 4.3 currently undermines the central claim of unique behavior variance.

major comments (4)
  1. [§4.3, Figure 5] The top-2 Text representations (CBOW GoogleNews, fastText CommonCrawl) and the top Behavior representation (PPMI SVD SWOW) are selected based on overall RCA performance on the same 292 norms that are then used to compute the ensemble differences. This selection-on-evaluation inflates the reported unique-variance gains on Dominance, Arousal, Valence, Emotion, Goals/Needs, Motor, and Social/Moral, because the representations were chosen to perform well on these very targets. The Wilcoxon signed-rank tests reported in Figure 5 compare within the selected representations only and do not account for the selection step. Please re-run the analysis with a hold-out-norm or split-half selection procedure, or otherwise provide evidence that the category-level advantages are not driven by selection noise.
  2. [§4.2, Table 1] The 'norms sensorimotor' behavior representation is the same rating data (Lynott et al., 2020) that constitutes the Sensory and Motor norm categories in the psychNorms metabase. Probing this representation against those norms is circular and can drive the apparent behavior advantages on Motor and related categories. The paper should exclude this representation from RCA when its own norms serve as targets, or explicitly report results without it. This is especially important because Section 4.2's claim that 'the best-performing behavior representations perform comparatively strongly on ... Motor' may be entirely due to this circular case.
  3. [§4.3] The Text & Text ensemble concatenates two text representations, while the Text & Behavior ensemble concatenates one text representation and one behavior representation. The feature dimensionality of these concatenated spaces is not matched, and linear probe performance can depend on the number of input features even with L2 regularization (the optimal regularization path and the effective model capacity change). The reported marginal R2 gains may therefore reflect the dimensionality difference rather than the unique semantic content of behavior features. Please match the dimensionality across ensembles (e.g., by subsampling features or using a fixed total dimension) or show that the results are robust to the feature-count difference.
  4. [§4.2] The behavior representations (e.g., free associations) and the norm ratings are both elicited from human participants. Common-method variance (e.g., shared response biases, word-frequency effects, or social desirability) could masquerade as 'unique psychological variance' attributed to behavior. The paper should at least discuss this alternative explanation and, ideally, include a control analysis using a behavior representation that is not based on human elicitation (e.g., eye-tracking) or partial out a general word-property factor.
minor comments (5)
  1. [§3.3] The text '1/apha' should read '1/alpha'.
  2. [§4.3] The typo 'Wilxocon' should be 'Wilcoxon'.
  3. [§4.1] The proportion of top-3 nearest neighbors for text is reported as '.97%' but the intended value is likely '97%'; please correct the formatting.
  4. [§4.3] The exact size of the collective vocabulary intersection used for the ensemble analysis is not reported; please provide it, as the probe sensitivity depends on this.
  5. [§4.3, Figure 5] The caption mentions 'all Text & Text and Text & Behavior ensemble combinations' but the main text does not specify how many combinations were evaluated or how they were aggregated; please clarify.

Circularity Check

1 steps flagged · score 2.0 of 10

One contained tautology: the 'norms sensorimotor' representation is the same human-rating data used as Sensory and Motor RCA targets, but the central Text & Behavior ensemble finding is independent.

  1. self definitional [Table 1 (norms sensorimotor row), Table 2 (Sensory and Motor categories), Section 3.3 (RCA definition), Figure 4]
    "Table 1 lists the behavior representation 'norms sensorimotor' as 'Ratings of 6 perceptual modalities and 5 action effectors (Lynott et al., 2020)', while Table 2 defines norm categories 'Sensory* Ratings of how strongly or easily the word is experienced through particular senses.' and 'Motor* Ratings of how much a word concerns bodily action or interaction.' Section 3.3 states: 'we fit a linear mapping ... to predict each norm (target) from each representation (features).'"

    The representation called 'norms sensorimotor' is literally the Lancaster Sensorimotor Norms by Lynott et al. (2020), and the Sensory and Motor norm targets in psychNorms are the same kind of human sensorimotor ratings (also Lynott et al., 2020). When RCA probes this representation on those targets, the feature matrix and the target vector are drawn from the same dataset, so a high pseudo-R2 on Sensory and Motor is a self-correlation rather than evidence that behavior-derived vectors encode psychological content. This tautology inflates the behavior group's profile in Figure 4 and Section 4.2, but it does not enter the Section 4.3 ensemble analysis, which uses the top behavior representation PPMI SVD SWOW (free associations) instead of norms sensorimotor.

full rationale

The paper's central claim is that behavior representations, specifically PPMI SVD SWOW trained on free associations, add unique variance beyond text on affective, agentic, and socio-moral norms. That claim is not circular: the free-association training data and the human rating norms are distinct measurement instruments, and the ensemble comparison in Section 4.3 does not use the 'norms sensorimotor' representation. The one clear by-construction overlap is the inclusion of 'norms sensorimotor' as a behavior representation while the same Lynott et al. sensorimotor ratings appear as Sensory and Motor norm targets; its performance there is tautological. The reviewer concern about selecting the top representations on the same 292 norms is a statistical validity threat (selection on the test set), not a circularity by the paper's own equations, so it does not affect the circularity score. No load-bearing self-citation chain was found. Score 2 reflects one contained, non-central tautology.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim is empirical and depends on the probe methodology, the validity of the norm ratings, and the specific choice of representations for the ensemble comparison. I list the hand-chosen threshold and category grouping as free parameters, and the methodological assumptions as axioms. No new theoretical entities are postulated.

free parameters (2)
  • Minimum probe test-set size threshold of 20 samples = 20
    Per Section 3.3, cases with fewer than 20 samples in the intersection vocabulary are not probed. This hand-chosen threshold affects which norms are included in category averages, and the paper notes that in a minority of cases category averages are estimated from a reduced set of norms.
  • Norm category grouping = 27 categories
    The 292 norms are manually grouped into 27 higher-level categories (Table 2, Section 3.1). The central claims are stated at the category level (affective, agentic, socio-moral), so the grouping determines the headline results.
assumptions (5)
  • domain assumption Linear probes faithfully reflect the information content of a representation (Hupkes et al. 2018).
    Invoked in Section 3.3 to justify using L2-regularized linear classifiers/regressors to measure norm content. If linear probes miss nonlinear encodings, RCA underestimates content; if they overfit, it overestimates.
  • standard math RSA with cosine similarity and Spearman correlation on the upper triangle captures meaningful representational similarity.
    Used in Section 3.2 following Kriegeskorte et al. 2008 and Lenci et al. 2022. Standard in the field.
  • domain assumption Word norm ratings are valid measures of psychological content (affect, agency, socio-moral dimensions).
    The RCA result (which norms are encoded) is only as meaningful as the psychNorms ratings. The paper relies on the validity of the 292 norms from the literature.
  • ad hoc to paper The 'top' representations for the ensemble analysis were selected on overall RCA performance, which does not inflate the estimate of unique variance.
    Section 4.3 selects CBOW GoogleNews, fastText CommonCrawl, and PPMI SVD SWOW as top performers based on Section 4.2 results on the same 292 norms. This selection-on-the-test-set can inflate the observed marginal gain from ensembling behavior with text.
  • domain assumption The common-vocabulary subsetting in Section 4.3 does not bias the comparison between ensembles.
    Subsetting all representations to their collective intersection makes training sets identical across conditions, but reduces the vocabulary and can reduce sensitivity differently for representations with different vocabulary distributions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Probing the contents of semantic representations from text, behavior, and brain data using the psychNorms metabase." pith.science (2026). https://pith.science/paper/CBG7VD43

@misc{pith2026241204936,
  author       = {Pith},
  title        = {Pith review of: Probing the contents of semantic representations from text, behavior, and brain data using the psychNorms metabase},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CBG7VD43}},
  note         = {Machine review of arXiv:2412.04936}
}
read the original abstract

Semantic representations are integral to natural language processing, psycholinguistics, and artificial intelligence. Although often derived from internet text, recent years have seen a rise in the popularity of behavior-based (e.g., free associations) and brain-based (e.g., fMRI) representations, which promise improvements in our ability to measure and model human representations. We carry out the first systematic evaluation of the similarities and differences between semantic representations derived from text, behavior, and brain data. Using representational similarity analysis, we show that word vectors derived from behavior and brain data encode information that differs from their text-derived cousins. Furthermore, drawing on our psychNorms metabase, alongside an interpretability method that we call representational content analysis, we find that, in particular, behavior representations capture unique variance on certain affective, agentic, and socio-moral dimensions. We thus establish behavior as an important complement to text for capturing human representations and behavior. These results are broadly relevant to research aimed at learning human-aligned semantic representations, including work on evaluating and aligning large language models.

Figures

Figures reproduced from arXiv: 2412.04936 by the authors.

Figure 1
Figure 1. An illustration of our approach. Word vectors are first obtained from the different data sources and then [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. An illustration of the size of the vocabularies (y-axis, log-scaled) for each representation and norm (x-axis, [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. A: A 2-dimensional projection of the representational similarity space. The space was obtained by multidimensional scaling of the pairwise Spearman dissimilarity matrix between representations. Text = green, behavior = purple, brain = blue. B: A heatmap visualization of the pairwise Spearman similarity matrix. representational similarity analysis (RSA), we com￾pare the representations obtained from each data type (s… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Average 5-fold cross-validation (pseudo-) [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Average 5-fold cross-validation (pseudo-) [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Addressing Longstanding Challenges in Cognitive Science with Language Models

    cs.AI 2025-10 conditional novelty 4.0 of 10

    A review proposes that LLMs can serve as tools for a more integrative and cumulative cognitive science when used under human oversight.

Reference graph

Works this paper leans on

54 extracted references · 34 canonical work pages · cited by 1 Pith paper

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Katherine Abramski, Clara Lavorati, Giulio Rossetti, and Massimo Stella. 2024. LLM -generated word association norms. In HHAI 2024: Hybrid Human AI Systems for the Social Good , pages 3--12. IOS Press

  4. [4]

    Samuel Aeschbach, Rui Mata, and Dirk U Wulff. 2024. https://doi.org/10.31234/osf.io/ra87s Mapping the mind with free associations: A tutorial using the R package associator . OSF

  5. [5]

    Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, et al. 2022. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862

  6. [6]

    Lisa Beinborn, Samira Abnar, and Rochelle Choenni. 2019. Robust evaluation of language--brain encoding experiments. In International Conference on Computational Linguistics and Intelligent Text Processing, pages 44--61. Springer

  7. [7]

    Yonatan Belinkov. 2022. Probing classifiers: Promises, shortcomings, and advances. Computational Linguistics, 48(1):207--219

  8. [8]

    Jeffrey R Binder, Lisa L Conant, Colin J Humphries, Leonardo Fernandino, Stephen B Simons, Mario Aguilar, and Rutvik H Desai. 2016. Toward a brain-based componential semantic representation. Cognitive neuropsychology, 33(3-4):130--174

Show all 54 references
  1. [9]

    Gemma Boleda. 2020. Distributional semantics and linguistic theory. Annual Review of Linguistics, 6(1):213--234

  2. [10]

    Jonathan R Brennan, Edward P Stabler, Sarah E Van Wagenen, Wen-Ming Luh, and John T Hale. 2016. Abstract linguistic structure correlates with temporal activity during naturalistic comprehension. Brain and language, 157:81--94

  3. [11]

    Michael P Broderick, Andrew J Anderson, Giovanni M Di Liberto, Michael J Crosse, and Edmund C Lalor. 2018. Electrophysiological correlates of semantic dissimilarity reflect the comprehension of natural, narrative speech. Current Biology, 28(5):803--809

  4. [12]

    Marc Brysbaert and Boris New. 2009. Moving beyond ku c era and francis: A critical evaluation of current word frequency norms and the introduction of a new and improved word frequency measure for american english. Behavior research methods, 41(4):977--990

  5. [13]

    Erin M Buchanan, Kathrene D Valentine, and Nicholas P Maxwell. 2019. English semantic feature production norms: An extended database of 4436 concepts. Behavior Research Methods, 51:1849--1863

  6. [14]

    Daniel Cer, Mona Diab, Eneko Agirre, Inigo Lopez-Gazpio, and Lucia Specia. 2017. Semeval-2017 task 1: Semantic textual similarity-multilingual and cross-lingual focused evaluation. arXiv preprint arXiv:1708.00055

  7. [15]

    Alexis Conneau, German Kruszewski, Guillaume Lample, Lo \" c Barrault, and Marco Baroni. 2018. What you can cram into a single vector: Probing sentence embeddings for linguistic properties. arXiv preprint arXiv:1805.01070

  8. [16]

    Common Crawl. 2024. Common crawl. http://commoncrawl.org. Accessed: 2024-11-20

  9. [17]

    Small World of Words

    Simon De Deyne, Danielle J Navarro, Amy Perfors, Marc Brysbaert, and Gert Storms. 2019. The “ Small World of Words ” english word association norms for over 12,000 cue words. Behavior research methods, 51:987--1006

  10. [18]

    Chuanji Gao, Svetlana V Shinkareva, and Rutvik H Desai. 2023. Scope: The south carolina psycholinguistic metabase. Behavior research methods, 55(6):2853--2884

  11. [19]

    Yoav Goldberg and Omer Levy. 2014. word2vec explained: deriving mikolov et al.'s negative-sampling word-embedding method. arXiv preprint arXiv:1402.3722

  12. [20]

    Fritz G \"u nther, Luca Rinaldi, and Marco Marelli. 2019. Vector-space models of semantic representation from a cognitive perspective: A discussion of common misconceptions. Perspectives on Psychological Science, 14(6):1006--1033

  13. [21]

    Martin N Hebart, Oliver Contier, Lina Teichmann, Adam H Rockter, Charles Y Zheng, Alexis Kidder, Anna Corriveau, Maryam Vaziri-Pashkam, and Chris I Baker. 2023. Things-data, a multimodal collection of large-scale datasets for investigating object representations in human brain...

  14. [22]

    Martin N Hebart, Charles Y Zheng, Francisco Pereira, and Chris I Baker. 2020. Revealing the multidimensional mental representations of natural objects underlying human similarity judgements. Nature human behaviour, 4(11):1173--1185

  15. [23]

    Andrew C Heusser, Kirsten Ziman, Lucy LW Owen, and Jeremy R Manning. 2017. Hypertools: A Python toolbox for visualizing and manipulating high-dimensional data. arXiv preprint arXiv:1701.08290

  16. [24]

    Nora Hollenstein, Antonio de la Torre, Nicolas Langer, and Ce Zhang. 2019. CogniVal : A framework for cognitive word embedding evaluation. arXiv preprint arXiv:1909.09001

  17. [25]

    Nora Hollenstein, Jonathan Rotsztejn, Marius Troendle, Andreas Pedroni, Ce Zhang, and Nicolas Langer. 2018. ZuCo , a simultaneous EEG and eye-tracking resource for natural sentence reading. Scientific data, 5(1):1--13

  18. [26]

    Dieuwke Hupkes, Sara Veldhoen, and Willem Zuidema. 2018. Visualisation and'diagnostic classifiers' reveal how recurrent and recursive neural networks process hierarchical structure. Journal of Artificial Intelligence Research, 61:907--926

  19. [27]

    Zak Hussain, Marcel Binz, Rui Mata, and Dirk U Wulff. 2024 a . A tutorial on open-source large language models for behavioral science. Behavior Research Methods, pages 1--24

  20. [28]

    Zak Hussain, Rui Mata, and Dirk U Wulff. 2024 b . Novel embeddings improve the prediction of risk perception. EPJ data science, 13(1):38

  21. [29]

    Mohsen Jamali, Benjamin Grannan, Jing Cai, Arjun R Khanna, William Mu \ n oz, Irene Caprara, Angelique C Paulk, Sydney S Cash, Evelina Fedorenko, and Ziv M Williams. 2024. Semantic encoding during language comprehension at single-cell resolution. Nature, pages 1--7

  22. [30]

    Daniel Jurafsky and James H. Martin. 2024. https://web.stanford.edu/ jurafsky/slp3/ Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition with Language Models , 3rd edition. Online manuscript released ...

  23. [31]

    Casey Kennington. 2021. Enriching language models with visually-grounded word vectors and the lancaster sensorimotor norms. In Proceedings of the 25th Conference on Computational Natural Language Learning, pages 148--157

  24. [32]

    Kiss, Christine Armstrong, Robert Milroy, and James Piper

    George R. Kiss, Christine Armstrong, Robert Milroy, and James Piper. 1973. An associative thesaurus of english and its computer analysis. In The Computer and Literary Studies, pages 153--165. Edinburgh University Press, Edinburgh, UK

  25. [33]

    Nikolaus Kriegeskorte, Marieke Mur, and Peter A Bandettini. 2008. Representational similarity analysis-connecting the branches of systems neuroscience. Frontiers in systems neuroscience, page 4

  26. [34]

    Alessandro Lenci, Magnus Sahlgren, Patrick Jeuniaux, Amaru Cuba Gyllensten, and Martina Miliani. 2022. A comparative evaluation and analysis of three generations of distributional semantic models. Language resources and evaluation, 56(4):1269--1313

  27. [35]

    Minh-Thang Luong, Richard Socher, and Christopher D Manning. 2013. Better word representations with recursive neural networks for morphology. In Proceedings of the seventeenth conference on computational natural language learning, pages 104--113

  28. [36]

    Dermot Lynott, Louise Connell, Marc Brysbaert, James Brand, and James Carney. 2020. The Lancaster Sensorimotor Norms : multidimensional measures of perceptual and action strength for 40,000 english words. Behavior research methods, 52:1271--1291

  29. [37]

    Yu Meng, Jiaxin Huang, Guangyuan Wang, Chao Zhang, Honglei Zhuang, Lance Kaplan, and Jiawei Han. 2019. Spherical text embedding. In Advances in neural information processing systems

  30. [38]

    Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. https://doi.org/10.48550/arXiv.1301.3781 Efficient estimation of word representations in vector space . arXiv preprint

  31. [39]

    Tomas Mikolov, Edouard Grave, Piotr Bojanowski, Christian Puhrsch, and Armand Joulin. 2018. Advances in pre-training distributed word representations. In Proceedings of the International Conference on Language Resources and Evaluation (LREC 2018)

  32. [40]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. 2011. Scikit-learn: Machine learning in P ython. Journal of Machine L...

  33. [41]

    Jeffrey Pennington, Richard Socher, and Christopher D Manning. 2014. GloVe : Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pages 1532--1543

  34. [42]

    Marcus E Raichle. 2010. Two views of brain function. Trends in cognitive sciences, 14(4):180--190

  35. [43]

    Russell Richie and Sudeep Bhatia. 2021. Similarity judgment within and across categories: A comprehensive model comparison. Cognitive science, 45(8):e13030

  36. [44]

    o zde G \

    G \"o zde G \"u l S ahin, Clara Vania, Ilia Kuznetsov, and Iryna Gurevych. 2020. https://doi.org/10.1162/coli_a_00376 LINSPECTOR : Multilingual probing tasks for word representations . Computational Linguistics, 46(2):335--385

  37. [45]

    Alexandre Salle, Marco Idiart, and Aline Villavicencio. 2016. Matrix factorization using window sampling and negative sampling for improved word representations. arXiv preprint arXiv:1606.00819

  38. [46]

    Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D Manning, Andrew Y Ng, and Christopher Potts. 2013. Recursive deep models for semantic compositionality over a sentiment treebank. In Proceedings of the 2013 conference on empirical methods in natural language...

  39. [47]

    Ilia Sucholutsky, Lukas Muttenthaler, Adrian Weller, Andi Peng, Andreea Bobu, Been Kim, Bradley C Love, Erin Grant, Jascha Achterberg, Joshua B Tenenbaum, et al. 2023. Getting aligned on representational alignment. arXiv preprint arXiv:2310.13018

  40. [48]

    Avijit Thawani, Biplav Srivastava, and Anil Singh. 2019. Swow-8500: Word association task for intrinsic evaluation of word embeddings. In Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for NLP, pages 43--51

  41. [49]

    Mariya Toneva and Leila Wehbe. 2019. Interpreting and improving natural-language processing (in machines) with natural language-processing (in the brain). Advances in neural information processing systems, 32

  42. [50]

    Jeroen Van Paridon and Bill Thompson. 2021. subs2vec: Word embeddings from subtitles in 55 languages. Behavior research methods, 53(2):629--655

  43. [51]

    Sudhir Varma and Richard Simon. 2006. https://doi.org/10.1186/1471-2105-7-91 Bias in error estimation when using cross-validation for model selection . BMC Bioinformatics, 7(1):1--8

  44. [52]

    Leila Wehbe, Brian Murphy, Partha Talukdar, Alona Fyshe, Aaditya Ramdas, and Tom Mitchell. 2014. Simultaneously uncovering the patterns of brain regions involved in different story reading subprocesses. PloS one, 9(11):e112575

  45. [53]

    Dirk U Wulff and Rui Mata. 2022. On the semantic representation of risk. Science advances, 8(27):eabm1883

  46. [54]

    Nicolas Yax, Hernan Anll \'o , and Stefano Palminteri. 2024. Studying and improving reasoning in humans and machines. Communications Psychology, 2(1):51

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.