Interpretable machine learning for psychological research: Opportunities and pitfalls
In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are typically not only interested in making predictions about the dependent variable, but also in learning which predictor variables are relevant,...
Uložené v:
| Vydané v: | Psychological methods Ročník 30; číslo 2; s. 271 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
United States
01.04.2025
|
| Predmet: | |
| ISSN: | 1939-1463, 1939-1463 |
| On-line prístup: | Zistit podrobnosti o prístupe |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are typically not only interested in making predictions about the dependent variable, but also in learning which predictor variables are relevant, how they influence the dependent variable, and which predictors interact with each other. However, most machine learning methods are not directly interpretable. Interpretation techniques that support researchers in describing how the machine learning technique came to its prediction may be a means to this end. We present a variety of interpretation techniques and illustrate the opportunities they provide for interpreting the results of two widely used black box machine learning methods that serve as our examples: random forests and neural networks. At the same time, we illustrate potential pitfalls and risks of misinterpretation that may occur in certain data settings. We show in which way correlated predictors impact interpretations with regard to the relevance or shape of predictor effects and in which situations interaction effects may or may not be detected. We use simulated didactic examples throughout the article, as well as an empirical data set for illustrating an approach to objectify the interpretation of visualizations. We conclude that, when critically reflected, interpretable machine learning techniques may provide useful tools when describing complex psychological relationships. (PsycInfo Database Record (c) 2025 APA, all rights reserved). |
|---|---|
| AbstractList | In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are typically not only interested in making predictions about the dependent variable, but also in learning which predictor variables are relevant, how they influence the dependent variable, and which predictors interact with each other. However, most machine learning methods are not directly interpretable. Interpretation techniques that support researchers in describing how the machine learning technique came to its prediction may be a means to this end. We present a variety of interpretation techniques and illustrate the opportunities they provide for interpreting the results of two widely used black box machine learning methods that serve as our examples: random forests and neural networks. At the same time, we illustrate potential pitfalls and risks of misinterpretation that may occur in certain data settings. We show in which way correlated predictors impact interpretations with regard to the relevance or shape of predictor effects and in which situations interaction effects may or may not be detected. We use simulated didactic examples throughout the article, as well as an empirical data set for illustrating an approach to objectify the interpretation of visualizations. We conclude that, when critically reflected, interpretable machine learning techniques may provide useful tools when describing complex psychological relationships. (PsycInfo Database Record (c) 2025 APA, all rights reserved). In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are typically not only interested in making predictions about the dependent variable, but also in learning which predictor variables are relevant, how they influence the dependent variable, and which predictors interact with each other. However, most machine learning methods are not directly interpretable. Interpretation techniques that support researchers in describing how the machine learning technique came to its prediction may be a means to this end. We present a variety of interpretation techniques and illustrate the opportunities they provide for interpreting the results of two widely used black box machine learning methods that serve as our examples: random forests and neural networks. At the same time, we illustrate potential pitfalls and risks of misinterpretation that may occur in certain data settings. We show in which way correlated predictors impact interpretations with regard to the relevance or shape of predictor effects and in which situations interaction effects may or may not be detected. We use simulated didactic examples throughout the article, as well as an empirical data set for illustrating an approach to objectify the interpretation of visualizations. We conclude that, when critically reflected, interpretable machine learning techniques may provide useful tools when describing complex psychological relationships. (PsycInfo Database Record (c) 2023 APA, all rights reserved).In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are typically not only interested in making predictions about the dependent variable, but also in learning which predictor variables are relevant, how they influence the dependent variable, and which predictors interact with each other. However, most machine learning methods are not directly interpretable. Interpretation techniques that support researchers in describing how the machine learning technique came to its prediction may be a means to this end. We present a variety of interpretation techniques and illustrate the opportunities they provide for interpreting the results of two widely used black box machine learning methods that serve as our examples: random forests and neural networks. At the same time, we illustrate potential pitfalls and risks of misinterpretation that may occur in certain data settings. We show in which way correlated predictors impact interpretations with regard to the relevance or shape of predictor effects and in which situations interaction effects may or may not be detected. We use simulated didactic examples throughout the article, as well as an empirical data set for illustrating an approach to objectify the interpretation of visualizations. We conclude that, when critically reflected, interpretable machine learning techniques may provide useful tools when describing complex psychological relationships. (PsycInfo Database Record (c) 2023 APA, all rights reserved). |
| Author | Rothacher, Yannick Henninger, Mirka Debelak, Rudolf Strobl, Carolin |
| Author_xml | – sequence: 1 givenname: Mirka orcidid: 0000-0003-4676-2361 surname: Henninger fullname: Henninger, Mirka organization: Institute of Psychology, University of Zurich – sequence: 2 givenname: Rudolf surname: Debelak fullname: Debelak, Rudolf organization: Institute of Psychology, University of Zurich – sequence: 3 givenname: Yannick surname: Rothacher fullname: Rothacher, Yannick organization: Institute of Psychology, University of Zurich – sequence: 4 givenname: Carolin surname: Strobl fullname: Strobl, Carolin organization: Institute of Psychology, University of Zurich |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/37227894$$D View this record in MEDLINE/PubMed |
| BookMark | eNpNkD1PwzAYhC1URGlh4QcgjywBx3bimA1VfFSq1IXO0VvnTWvkOMF2hv57IlEkbnlOutMNtyAz33sk5C5njzkT6qnDxCYVJbsg17kWOstlKWb__JwsYvxiLJeikldkLhTnqtLymuzWPmEYAibYO6QdmKP1SB1C8NYfaNsHOsSTOfauP1gDjgaMU2iOz3Q7DH1Io7fJYqTgGzrY1IJz8YZcTox4e-aS7N5eP1cf2Wb7vl69bDKQXKRMG8WkwL2SbWuYAcVa1rQFllBBoRteNIBlqwwYw4XOEbTJFaBsFNe8UMCX5OF3dwj994gx1Z2NBp0Dj_0Ya15xxgQTUk3V-3N13HfY1EOwHYRT_XcF_wHmOmO0 |
| CitedBy_id | crossref_primary_10_1007_s41237_024_00252_3 crossref_primary_10_1007_s41237_024_00253_2 crossref_primary_10_5093_clh2025a13 crossref_primary_10_1057_s41267_024_00687_6 crossref_primary_10_1515_cllt_2024_0028 crossref_primary_10_3758_s13428_024_02588_w crossref_primary_10_1177_25152459251351285 crossref_primary_10_1177_25152459251345696 crossref_primary_10_1080_13803395_2025_2458547 crossref_primary_10_3102_10769986231193327 crossref_primary_10_1111_bmsp_12375 crossref_primary_10_15626_MP_2023_3796 crossref_primary_10_1080_17437199_2024_2400977 crossref_primary_10_1007_s10489_024_06130_5 crossref_primary_10_1111_bmsp_70009 crossref_primary_10_1007_s10488_023_01328_0 crossref_primary_10_1038_s41398_025_03360_0 crossref_primary_10_1002_ab_70013 crossref_primary_10_1177_10944281251323248 crossref_primary_10_3390_make6020061 |
| ContentType | Journal Article |
| DBID | CGR CUY CVF ECM EIF NPM 7X8 |
| DOI | 10.1037/met0000560 |
| DatabaseName | Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic |
| DatabaseTitle | MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic |
| DatabaseTitleList | MEDLINE MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| Discipline | Psychology |
| EISSN | 1939-1463 |
| ExternalDocumentID | 37227894 |
| Genre | Journal Article |
| GroupedDBID | --- --Z -~X .-4 07C 0R~ 123 29P 354 3KI 53G 5VS 7RZ ABIVO ABNCP ABVOZ ACHQT ACPQG AEHFB AETEA AFFHD ALMA_UNASSIGNED_HOLDINGS AWKKM AZXWR CGNQK CGR CS3 CUY CVF ECM EIF EPA F5P FTD HVGLF HZ~ ISO LW5 NPM O9- OHT OPA OVD P2P PHGZT ROL SES SPA TEORI TN5 UHS YNT ZPI 7X8 PUEGO |
| ID | FETCH-LOGICAL-a423t-9c7043eb74ffc0ca70f0df5e6a8a59d25dae6f7cacc2391ea9c17ae4d729257a2 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 34 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000995142400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1939-1463 |
| IngestDate | Sun Sep 28 08:47:21 EDT 2025 Fri Nov 21 01:40:50 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 2 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-a423t-9c7043eb74ffc0ca70f0df5e6a8a59d25dae6f7cacc2391ea9c17ae4d729257a2 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0003-4676-2361 |
| PMID | 37227894 |
| PQID | 2820030347 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_2820030347 pubmed_primary_37227894 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-04-01 |
| PublicationDateYYYYMMDD | 2025-04-01 |
| PublicationDate_xml | – month: 04 year: 2025 text: 2025-04-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | Psychological methods |
| PublicationTitleAlternate | Psychol Methods |
| PublicationYear | 2025 |
| SSID | ssj0014384 |
| Score | 2.5822637 |
| Snippet | In recent years, machine learning methods have become increasingly popular prediction methods in psychology. At the same time, psychological researchers are... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| StartPage | 271 |
| SubjectTerms | Data Interpretation, Statistical Humans Machine Learning Neural Networks, Computer Psychology - methods |
| Title | Interpretable machine learning for psychological research: Opportunities and pitfalls |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/37227894 https://www.proquest.com/docview/2820030347 |
| Volume | 30 |
| WOSCitedRecordID | wos000995142400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV27TsMwFLWAMnTh_SgvGYnVIomd2GZBCFExQOlAUbfI8QMqlTSQFIm_x3ZdysCAxJIpkeKbm-Nj3-N7ADjLhFAmlgoRnhFEVJEixrzWNRVcRVIbLx5_uqO9HhsOeT9suNVBVjnHRA_UaiLdHvm5XRq4hMSEXlZvyLlGuepqsNBYBi1sqYyTdNHhoopAMAtVZY4sIuB5e1Jsl_u6cVCdhtaUv1JLP8V01__7chtgLZBLeDXLhk2wpMst0P7GuM9tMFiIDIuxhq9eSqlh8I54hpbCwuonJsLQDejlAj5UjqtPS9-DFYpSwWrUGDEe1ztg0L15vL5FwVkBCUufGsQljQjWBSXGyEgKGplImVRngomUqyRVQmeGSiFlgnmsBZcxFZooS8XtPy6SXbBSTkq9DyAzrBBF5m6w5MZ-A66ZSYmMueEFL-IOOJ2HLLeZ68oRotSTaZ0vgtYBe7O459WsxUaOqT-iSw7-8PQhaCfOlNfLaY5Ayw681sdgVX40o_r9xKeEvfb691_aC8PY |
| linkProvider | ProQuest |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Interpretable+machine+learning+for+psychological+research%3A+Opportunities+and+pitfalls&rft.jtitle=Psychological+methods&rft.au=Henninger%2C+Mirka&rft.au=Debelak%2C+Rudolf&rft.au=Rothacher%2C+Yannick&rft.au=Strobl%2C+Carolin&rft.date=2025-04-01&rft.issn=1939-1463&rft.eissn=1939-1463&rft_id=info:doi/10.1037%2Fmet0000560&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1939-1463&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1939-1463&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1939-1463&client=summon |