Family lexicon: Using language models to encode memories of personally familiar and famous people and places in the brain

Andrea Bruera; Massimo Poesio

doi:10.1371/journal.pone.0291099

Peer Review History

Original SubmissionAugust 31, 2023
14 May 2024 Decision Letter - Michal Ptaszynski, Editor PONE-D-23-26777Family Lexicon: using language models to encode memories of personally familiar and famous people and places in the brainPLOS ONE Dear Dr. Bruera, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Jun 28 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Michal Ptaszynski, PhD Academic Editor PLOS ONE Journal requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. Please include a complete copy of PLOS’ questionnaire on inclusivity in global research in your revised manuscript. Our policy for research in this area aims to improve transparency in the reporting of research performed outside of researchers’ own country or community. The policy applies to researchers who have travelled to a different country to conduct research, research with Indigenous populations or their lands, and research on cultural artefacts. The questionnaire can also be requested at the journal’s discretion for any other submissions, even if these conditions are not met. Please find more information on the policy and a link to download a blank copy of the questionnaire here: https://journals.plos.org/plosone/s/best-practices-in-research-reporting. Please upload a completed version of your questionnaire as Supporting Information when you resubmit your manuscript. 3. Note from Emily Chenette, Editor in Chief of PLOS ONE, and Iain Hrynaszkiewicz, Director of Open Research Solutions at PLOS: Did you know that depositing data in a repository is associated with up to a 25% citation advantage (https://doi.org/10.1371/journal.pone.0230416)? If you’ve not already done so, consider depositing your raw data in a repository to ensure your work is read, appreciated and cited by the largest possible audience. You’ll also earn an Accessible Data icon on your published paper if you deposit your data in any participating repository (https://plos.org/open-science/open-data/#accessible-data). 4. Thank you for stating the following financial disclosure: [AB was supported by a doctoral studentship from the School of Electronic Engineering and Computer Science, Queen Mary University of London.]. Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." If this statement is not correct you must amend it as needed. Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf. 5. We note that you have indicated that there are restrictions to data sharing for this study. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Before we proceed with your manuscript, please address the following prompts: a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., a Research Ethics Committee or Institutional Review Board, etc.). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent. b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of recommended repositories, please see https://journals.plos.org/plosone/s/recommended-repositories. You also have the option of uploading the data as Supporting Information files, but we would recommend depositing data directly to a data repository if possible. We will update your Data Availability statement on your behalf to reflect the information you provide. 6. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well. 7. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: Yes Reviewer #4: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: Yes Reviewer #4: Yes ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: No Reviewer #2: Yes Reviewer #3: No Reviewer #4: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: Yes Reviewer #4: Yes ******** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: The research domain and rationale are clearly defined, thanks to a rich and detailed Introduction section. The overall objective proposes a contextualised, large-language model-based approach to capture fine-grained personally-familiar entities (places and expecially persons) better than more tradiotonal static language models (such as those using word2vec representations). Results support the suggested approach, in particular for what it concerns personally familiar people; whereas worse results are shown for personally familiar places. In such account, the proposed explanation is not totally convincing. My suggestion is to clearly state that more analyses should be done to clarify to clarify the differences in encoding places versus people. If the Introduction section is clearly written, the other sections need small revisions from yours: - line 115: Fig 1 left portion --> right portion - line 116: We --> we - lines 116-117: .. had not either ... or --> had neither .. nor - line 128: pleade add standard deviations as well - line132 and all ensuing occurrences: see Section --> reference to Section is always missing - line 248: (Wikipedia or questionnairre) --> why Wikipedia? in lines 155-157 I read that for individual entity you take names and a short description (occupations for people and type of places) - lines 251-254: here the syntactic structure is a bit odd and access to the content is quite difficult. - line 274: at at (remove one of them) - line 291: odd punctuation (sentence; and so on --> sentence, and so on) - lines 295-302: pay attention to the consecutio temporum. Reviewer #2: Regrettably on reading the full paper I discovered that I am not competent to review it becaue there is virtually no linguistics in it but computational encephelography. I apologize for any inconveneince caused by my inept decision. Although I have recommended rejection it was a forced decision and should be ignored. Reviewer #3: Thank you for letting me review this very interesting paper on how language models can encode subject-specific information. The topic itself is very interesting because usually language models try to capture concepts from large amounts of data, while in this case, we see how they can also be used to encode some fragments of personal life. The manuscript is well written and the study is properly evaluated. * Is the manuscript technically sound, and do the data support the conclusions? Yes. * Has the statistical analysis been performed appropriately and rigorously? The methodology used in the study is derived from the literature and appears soundly performed. * Have the authors made all data underlying the findings in their manuscript fully available? No, the data obtained from the study contain subject information that is not publicly available. However, the data used for the analysis have been made available. * Is the manuscript presented in an intelligible fashion and written in standard English? Yes, the manuscript is well-written. I have pointed out below a couple of sentences that could be improved for easier reading. Specific comments: Abstract: - I suggest to rewrite the sentence "In terms of spatio-temporal clusters, ... (500-700ms)." for better clarity. Methods -> Famous entities: - Lines 127-139: Have you checked that the assumptions of the t-test are met? Methods -> Personally familiar entities: - In line 154, you mention that you provided the text of the specific instructions given to the subjects in the code. Scrolling quickly through the repository I did not find it. I suggest that you 1) Update the Readme so that it is easier to find the information and 2) Provide a list of the questions/instructions as part of the manuscript (in the appendix?). I think it is a very interesting part of the study and that it should be included. Methods -> Spatio-temporal searchlight: - I suggest to rewrite the sentence in lines 371-373: "Results are intepreted as above... at a given point in time" for better clarity. Methods -> Statistical testing: - Line 380: We ran one-tailed statistical tests: which ones? Results: - It would be interesting to see the same figure as Figure 4, but for word2vec, so that it would also be possible to compare how this model performed for the different tasks. - Figure 3, Figure 4: it would be nice to add that the time radius for the searchlight was 50ms. - Figure 3, Figure 4: it would also be nice to add the x-axis under the searchlight coding. - Figure 4: What do the points indicated as stars in the graph represent? Other comments: - The hyperlinking of the sections did not work, see lines 132, 158, 178, 182, 240, 284, 497, 510,... - Line 204: "Each trial proceeded as followS". Missing S. - The acronym MNE is never spelled out. - Line 285: "pooling for the top four layerS". Missing S. - Figure 3 and Figure 4: space missing between "scalp locations" and "(searchlight)". Reviewer #4: In this paper, the authors use language models(both small and big model) to build a connection between semantic knowledge from personally familiar individual entities and semantic information in brain. And a series experiment show that encoding performance was significant in a large time window (200-800ms).I think the research topic is very interesting and owns practical value. There are some questions as follows, (1) why you choose 10 words to to form a topic? (2) It seems some hot words could own high weights and strong connections because experiment performance is better by big language than small model, have you try to test key-word's effect in your model? ******** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Claudia Marzi, ILC-CNR Reviewer #2: No Reviewer #3: No Reviewer #4: No ******** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0291099.r001
Revision 1
16 Aug 2024 Author Response To the attention of the Editorial Board of PLOS ONE, We are now ready to submit a revised version of our paper. We’d like to thank the editor, for arranging the reviewers' responses to our manuscript, and the reviewers, whose suggestions have helped us significantly in improving the quality of the paper and in making it clearer. We have taken into consideration all the comments, suggestions and concerns raised by the reviewers, and made the required revisions. We have reported all the individual answers to the reviewers in the next pages. In the response to the reviewers, the original reviewers’ comments are indicated by double quotation marks; our answer is in bold italics; the relevant sections from the revised manuscript are reported in italics (after the words 'NEW TEXT' or 'ADDED TEXT'). Summarizing the main changes, First of all, we have added a full figure (Figure 5), reporting complete, in-depth results for word2vec. This allows to see the differences between this model and the large language model, as suggested by a reviewer; Secondly, we have addressed the concerns put forward by two of the reviewers about the obscurity of some parts of the methodological details - namely related to the creation of the vectors for the entities, and to the experimental procedure. The relevant sections have been largely re-written; Thirdly, we have substituted t-tests - whose assumptions, as correctly pointed out by a reviewer, may not have been valid with our data - with non-parametric permutation tests, and corrected all relevant sections. We have also substituted the measure of Pearson correlation, that presented similar issues, with that of Spearman correlation, when comparing the model's predictions and the real EEG data; Then, we have explicitly acknowledged the limitations pointed out by one of the reviewers in the discussion; We have added as Supplementary Information a pdf containing a translation to English of the original questionnaire; Finally, we have added all details required by the reviewers to the figures, and we have changed all the colour palettes so as to make them readable also by people affected by colour blindness. In the version of the manuscript showing the changes between the first submission and the current revision, you will find the deleted text striked through and in blue; the added text in dark orange. We hope that you will find our responses satisfactory. Yours sincerely, Andrea Bruera & Massimo Poesio Response to the reviewers - in the following, we respond to all the points raised by the reviewers, reporting the original text preceded by an acronym for the reviewer (e.g. 'R1: ' for reviewer 1), and our answer preceded by 'answer: '. Reviewer #1 (R1): R1: "The research domain and rationale are clearly defined, thanks to a rich and detailed Introduction section. The overall objective proposes a contextualised, large-language model-based approach to capture fine-grained personally-familiar entities (places and especially persons) better than more traditional static language models (such as those using word2vec representations)." Answer: Thanks for your time and for raising very relevant points that, we believe, allowed us to noticeably improve the quality of this work. R1: "Results support the suggested approach, in particular for what it concerns personally familiar people; whereas worse results are shown for personally familiar places. In such account, the proposed explanation is not totally convincing. My suggestion is to clearly state that more analyses should be done to clarify to clarify the differences in encoding places versus people." Answer: Thanks for the suggestion. We have now added a paragraph where we discuss this point, and we acknowledged explicitly the current limitations. ADDED TEXT: "Turning to places, a more mixed set of results emerges. For XLM, a particularly striking difference in terms of encoding performance was found between personally familiar people and places (Fig 4): performance was barely above chance for places, while it was significant in a large time window for people. Interestingly, the patterns of encoding were quite similar for personally familiar people and places also for word2vec (Fig 5). At least two factors may be hypothesized to be at work. First of all, it’s possible that the questionnaire could not capture place-specific types of information - for instance, sensory and modality-specific features - that are crucial when it comes to cognitive processing of familiar places. Secondly, as discussed in [2, 60] , the identity of personally familiar places may be harder to process than that of personally familiar people - for various reasons that could be evolutionary, [132] social, [133] or related to the availability of semantic features during retrieval [134]. This would then make it harder to correctly distinguish place-specific, as opposed to person-specific, signatures in brain activity - a result that converges with the overall lower encoding scores for places reported here (Fig s 5 and Fig s 5) and that has previously been found in the literature [60, 134, 135] While we are unable to provide an answer to such questions with the current experiment, we believe this could be a fruitful direction for future research." R1: "If the Introduction section is clearly written, the other sections need small revisions from yours: - line 115: Fig 1 left portion --> right portion - line 116: We --> we - lines 116-117: .. had not either ... or --> had neither .. nor" Answer: Thanks for pointing out these typos, that we have now corrected. R1: "- line 128: pleade add standard deviations as well" Answer: We have now added all standard deviations. R1: "- line132 and all ensuing occurrences: see Section --> reference to Section is always missing" Answer: Sorry for the mistake. This should now have been fixed. R1: "- line 248: (Wikipedia or questionnairre) --> why Wikipedia? in lines 155-157 I read that for individual entity you take names and a short description (occupations for people and type of places)" Answer: Thanks for making us notice this point. We have now clarified when we used the texts (questionnaires or Wikipedia pages; to create the vectors) and when we employed categories (during the experiment). NEW TEXT: "For each famous individual entity we also collected the text from their Wikipedia page, under the assumption that such texts are a source of explicit knowledge regarding individual entities that can be mapped to the brain using their distributional information - an approach validated in neuroscience in [55] and in Natural Language Processing (NLP) in [41, 56, 57]. These texts will be used as described in Section 2.4.1 in order to extract semantic representations using language models. We also manually annotated for each famous person their occupation (e.g. politician, musician) and for each famous place its type (e.g. city, monument), since this information was used during the experimental task (see Section 2.2)" R1: "- lines 251-254: here the syntactic structure is a bit odd and access to the content is quite difficult." Answer: We have tried to clarify and simplify as much as possible the whole part dedicated to the vector extraction procedure. NEW TEXT: The rationale for this procedure is that words appearing in a text revolving around an entity carry a rich bundle of semantic information with respect to that entity. In other words we interpret the texts we collected as entity-specific distributional lexical information: for personally familiar people and places, the answers to the questionnaires; for their famous counterparts, the text from their Wikipedia pages [41, 55, 56]. The procedure was matched across models, so as to avoid methodological confounds. For each individual entity, we split the text in passages of at least 20 words. We set a lower passage length threshold because, when encoding sentences for downstream use, LLMs have been shown to work better with rather long passages [71]. In order to ensure that text portions were long enough to work well with XLM, we rearranged the text so that sequential passages of at least 20 words were created (i.e. if a sentence were shorter than 20 words, we considered to be part of the same passage as the following sentence). We chose 20 words as a threshold since in English sentence lengths are most commonly is between 10 and 30 words [72]). After having encoded all passages of text using the language models, we retained only the vectors in the sequences corresponding to content words (i.e. open-class words: nouns, verbs, adjectives, adverbs) from the corresponding descriptive text (Wikipedia or questionnaire). We decided to follow [44] and exclude closed-class words such as function words, since they do not carry semantic information. We reasoned that their presence would lead to the static language model being disadvantaged, since it has been shown that they struggle at representing function words properly [43, 73, 74]. Finally, we obtained a single entity representation in two steps. First, we averaged all the word vectors retained for each passage, thus obtaining one vector per passage. Then, we averaged all of the vectors for the individual passages of texts [32, 55] [75] . Therefore, at the end of the procedure we had one vector representation for each individual entity per model, capturing the distributional lexical information contained in our small-scale textual data. R1: "- line 274: at at (remove one of them) - line 291: odd punctuation (sentence; and so on --> sentence, and so on) - lines 295-302: pay attention to the consecutio temporum." Answer: Thanks - these should be fixed now. Reviewer #2 (R2): R2: "Regrettably on reading the full paper I discovered that I am not competent to review it because there is virtually no linguistics in it but computational encephalography. I apologize for any inconvenience caused by my inept decision. Although I have recommended rejection it was a forced decision and should be ignored." Answer: No worries, and thanks anyways for your help. Reviewer #3 (R3): R3: "Thank you for letting me review this very interesting paper on how language models can encode subject-specific information. The topic itself is very interesting because usually language models try to capture concepts from large amounts of data, while in this case, we see how they can also be used to encode some fragments of personal life. The manuscript is well written and the study is properly evaluated." Answer: Thanks for the insightful comments, suggestions and the positive feedback.. R3: "Specific comments: Abstract: - I suggest to rewrite the sentence "In terms of spatio-temporal clusters, ... (500-700ms)." for better clarity." Answer: Thanks for the suggestion. We have now fully reformulated the sentence and modified the paragraph, hopefully making it clearer. NEW TEXT: "Using spatio-temporal EEG searchlight, we found that we could predict brain responses significantly better than chance earlier (200-500ms) in bilateral temporo-parietal electrodes and later (500-700ms) in frontal and posterior central electrodes." R3: "Methods -> Famous entities: - Lines 127-139: Have you checked that the assumptions of the t-test are met?" Answer: We have now changed the analyses using nonparametric permutation tests, which make minimal assumptions about the underlying data distributions. R3: "Methods -> Personally familiar entities: - In line 154, you mention that you provided the text of the specific instructions given to the subjects in the code. Scrolling quickly through the repository I did not find it. I suggest that you 1) Update the Readme so that it is easier to find the information and 2) Provide a list of the questions/instructions as part of the manuscript (in the appendix?). I think it is a very interesting part of the study and that it should be included." Answer: Really sorry for this - we forgot to update the repository. We have now added the translated questionnaire as Supplementary Information and we have updated the repository containing the data and the code online. R3: "Methods -> Spatio-temporal searchlight: - I suggest to rewrite the sentence in lines 371-373: "Results are intepreted as above... at a given point in time" for better clarity." Answer: Thanks for pointing this out. We have now reformulated the sentence, with the aim of making it easier to understand. NEW TEXT: We were also interested in going beyond patterns of activity widely distributed across the whole scalp, and look at specific areas on the scalp where a model could explain evoked reponses. To this aim, we implemented a searchlight encoding analysis. Searchlight allows to find in a bottom-up fashion localized clusters of brain activity associated with an experimental condition, while exploiting the high sensitivity of multivariate analyses [91, 92]. In practice, searchlight consists of running the encoding analyses repeatedly across smaller clusters of electrodes on the scalp. To reduce the computational effort , we followed previous work [93] and used spatio-temporal clusters, where multiple time points are considered for each electrode within the cluster. As in [94], we employed a temporal radius of 50ms and a spatial radius of 30mm (i.e. a cluster contains evoked activity for 100ms, for electrodes falling within a circle having a diameter of 60mm). We computed statistical significance tests using the TFCE method described in Section 2.5.6. If the p-value for a cluster of electrodes at a given point in time fell below 0.05, we considered encoding to be significantly above chance. R3: "Methods -> Statistical testing: - Line 380: We ran one-tailed statistical tests: which ones?" Answer: Thanks for noticing this - we have now specified which statistical tests were used.. R3: "Results: - It would be interesting to see the same figure as Figure 4, but for word2vec, so that it would also be possible to compare how this model performed for the different tasks." Answer: Thanks for the suggestion. We have now added the full figure for word2vec (now Figure 5), and added the relevant parts to the Results section - which indeed makes clearer where the difference in performance between the two models stems from. NEW FIGURE: R3: "- Figure 3, Figure 4: it would be nice to add that the time radius for the searchlight was 50ms. - Figure 3, Figure 4: it would also be nice to add the x-axis under the searchlight coding." Answer: Thanks for both suggestions - we have now added the temporal radius in the figure, and the x-axis under the searchlight encoding results.. R3: "- Figure 4: What do the points indicated as stars in the graph represent?" Answer: Thanks a lot for spotting this! We had wrongly used a previous version of the figure, where stars represented points approaching significance ($p<.08$). Now only statistically significant ($p<0.05$) points are marked with a dot. R3: "Other comments: - The hyperlinking of the sections did not work, see lines 132, 158, 178, 182, 240, 284, 497, 510,... - Line 204: "Each trial proceeded as followS". Missing S. - The acronym MNE is never spelled out. - Line 285: "pooling for the top four layerS". Missing S. - Figure 3 and Figure 4: space missing between "scalp locations" and "(searchlight)"." Answer: Thanks for spotting these typos - they are now corrected. Reviewer #4 (R4): R2: "In this paper, the authors use language models(both small and big model) to build a connection between semantic knowledge from personally familiar individual entities and semantic information in brain. And a series experiment show that encoding performance was significant in a large time window (200-800ms).I think the research topic is very interesting and owns practical value. " Answer: Thanks for your feedback and the valuable comments. R4: "There are some questions as follows, (1) why you choose 10 words to to form a topic?" Answer: Thanks for the suggestion. We have now explicitly indicated from where we took each experimental parameter from. NEW TEXT: "We pre-trained a word2vec model o Attachments Attachment Submitted filename: letter editor bruera poesio plos one.pdf https://doi.org/10.1371/journal.pone.0291099.r002
17 Sep 2024 Decision Letter - Michal Ptaszynski, Editor Family lexicon: using language models to encode memories of personally familiar and famous people and places in the brain PONE-D-23-26777R1 Dear Dr. Bruera, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Michal Ptaszynski, PhD Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes ****** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: I strongly appreciate that the authors took into consideration all detailed comments from the reviewers, thus providing a revised version of their work which is highly relevant to a wide research community. Some minorities should be addressed: Line 279: there is a “is” that must be cancelled. Line 676: Fig s 3 and 5 Line 721: Fig s 5 and Fig s 5 ****** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Claudia Marzi ******** https://doi.org/10.1371/journal.pone.0291099.r003
Formally Accepted
2 Oct 2024 Acceptance Letter - Michal Ptaszynski, Editor PONE-D-23-26777R1 PLOS ONE Dear Dr. Bruera, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Michal Ptaszynski Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0291099.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .