{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T02:47:45Z","timestamp":1782096465455,"version":"3.54.5"},"posted":{"date-parts":[[2026,6,18]]},"group-title":"Health Informatics","reference-count":11,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T00:00:00Z","timestamp":1781740800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/publicdomain\/zero\/1.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,6,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>\n                  Previous work showed that many participants fail to notice a gorilla in a video of people playing basketball. Another study found that 83% of trained radiologists failed to report a gorilla figure inserted into a chest CT nodule-search task, even though eye-tracking revealed that most observers had foveated the figure. We ask whether a similar phenomenon exists in contemporary vision-language models (VLMs). We find that\n                  <jats:bold>(i)<\/jats:bold>\n                  VLMs are capable of spotting the gorilla in both still-frame images and videos of lung CT scans;\n                  <jats:bold>(ii)<\/jats:bold>\n                  models display inattentional blindness, which varies according to model generation and type of stimulus presented;\n                  <jats:bold>(iii)<\/jats:bold>\n                  Gemini-3.1-Pro outperforms most other flagship and open-weight VLMs at identifying the presence or absence of the gorilla. We additionally ran a segmentation experiment utilizing two different model classes: a generalist (SAM 3), which found the gorilla but produced little to no results for anatomy-based prompts; a medical specialist (BiomedParse), which produced more promising anatomy-based results but flagged \u201cgorilla\u201d on gorilla-free control videos on 82% of frames. The behavioral signature of inattentional blindness reproduces in VLMs, but a unique confabulation failure mode means that any \u201cdid the model see X\u201d claim requires signal-detection analysis with a matched-control false-alarm baseline.\n                <\/jats:p>","DOI":"10.64898\/2026.06.16.26355792","type":"posted-content","created":{"date-parts":[[2026,6,18]],"date-time":"2026-06-18T23:15:11Z","timestamp":1781824511000},"source":"Crossref","is-referenced-by-count":0,"title":["Looked but didn\u2019t see: inattentional blindness and yes-bias confabulation in vision-language models"],"prefix":"10.64898","author":[{"given":"Jonathan D.","family":"Raymond","sequence":"first","affiliation":[{"name":"Section on Emerging Analytics, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dat","family":"Duong","sequence":"additional","affiliation":[{"name":"Section on Emerging Analytics, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ping","family":"Hu","sequence":"additional","affiliation":[{"name":"Section on Emerging Analytics, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin D.","family":"Solomon","sequence":"additional","affiliation":[{"name":"Section on Emerging Analytics, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026062119100506000_2026.06.16.26355792v1.1","article-title":"SAM 3: Segment Anything with Concepts","year":"2026","journal-title":"International Conference on Learning Representations (ICLR)"},{"key":"2026062119100506000_2026.06.16.26355792v1.2","doi-asserted-by":"publisher","DOI":"10.1177\/0956797613479386"},{"key":"2026062119100506000_2026.06.16.26355792v1.3","doi-asserted-by":"publisher","DOI":"10.3758\/BF03203619"},{"key":"2026062119100506000_2026.06.16.26355792v1.4","doi-asserted-by":"publisher","DOI":"10.2307\/4615733"},{"key":"2026062119100506000_2026.06.16.26355792v1.5","doi-asserted-by":"publisher","DOI":"10.1145\/3748302"},{"key":"2026062119100506000_2026.06.16.26355792v1.6","doi-asserted-by":"crossref","unstructured":"Mack, A. , & Rock, I. (1998). Inattentional Blindness. MIT Press.","DOI":"10.7551\/mitpress\/3707.001.0001"},{"key":"2026062119100506000_2026.06.16.26355792v1.7","unstructured":"Macmillan, N. A. , & Creelman, C. D. (2005). Detection Theory: A User\u2019s Guide (2nd ed.). Lawrence Erlbaum Associates."},{"key":"2026062119100506000_2026.06.16.26355792v1.8","doi-asserted-by":"publisher","DOI":"10.1037\/0033-295X.112.1.217"},{"key":"2026062119100506000_2026.06.16.26355792v1.9","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1097-0258(19980430)17:8<873::AID-SIM779>3.0.CO;2-I"},{"key":"2026062119100506000_2026.06.16.26355792v1.10","doi-asserted-by":"publisher","DOI":"10.1068\/p2952"},{"issue":"1","key":"2026062119100506000_2026.06.16.26355792v1.11","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1038\/s41592-024-02499-w","article-title":"A foundation model for joint segmentation, detection and recognition of biomedical objects across nine modalities","volume":"22","year":"2025","journal-title":"Nature Methods"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.06.16.26355792","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T02:10:16Z","timestamp":1782094216000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.06.16.26355792"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,18]]},"references-count":11,"URL":"https:\/\/doi.org\/10.64898\/2026.06.16.26355792","relation":{},"subject":[],"published":{"date-parts":[[2026,6,18]]},"subtype":"preprint"}}