{"id":238469,"date":"2026-06-06T02:20:54","date_gmt":"2026-06-06T07:20:54","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2026\/06\/ai-fails-classic-attention-test-with-longer-word-lists-triggering-dramatic-accuracy-collapse"},"modified":"2026-06-06T02:20:54","modified_gmt":"2026-06-06T07:20:54","slug":"ai-fails-classic-attention-test-with-longer-word-lists-triggering-dramatic-accuracy-collapse","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2026\/06\/ai-fails-classic-attention-test-with-longer-word-lists-triggering-dramatic-accuracy-collapse","title":{"rendered":"AI fails classic attention test, with longer word lists triggering dramatic accuracy collapse"},"content":{"rendered":"<p><a class=\"aligncenter blog-photo\" href=\"https:\/\/lifeboat.com\/blog.images\/ai-fails-classic-attention-test-with-longer-word-lists-triggering-dramatic-accuracy-collapse2.jpg\"><\/a><\/p>\n<p>Giving AI a classic psychological test reveals an inherent weakness in LLM decision-making abilities. Suketu Patel and colleagues explored how transformer-based machine attention differs from human attention by testing AI models on the \u201cStroop task,\u201d in which words for colors are printed in colored ink, and participants are asked to name the ink color of each word while ignoring its meaning.<\/p>\n<p>The findings are <a href=\"https:\/\/academic.oup.com\/pnasnexus\/article\/doi\/10.1093\/pnasnexus\/pgag149\/8698838\" target=\"_blank\">published<\/a> in the journal PNAS Nexus.<\/p>\n<p>The task is clinically used to assess <a href=\"https:\/\/medicalxpress.com\/news\/2023-06-stop-signal-task-aid-adult-adhd.html?utm_source=embeddings&utm_medium=related&utm_campaign=internal\" rel=\"related\" target=\"_blank\">executive control<\/a>, especially a person\u2019s ability to inhibit an automatic response. Although humans generally take longer to answer correctly when words and colors are mismatched than when they match, they can still perform stably and with high accuracy even on long word lists.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Giving AI a classic psychological test reveals an inherent weakness in LLM decision-making abilities. Suketu Patel and colleagues explored how transformer-based machine attention differs from human attention by testing AI models on the \u201cStroop task,\u201d in which words for colors are printed in colored ink, and participants are asked to name the ink color of [\u2026]<\/p>\n","protected":false},"author":427,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11,6],"tags":[],"class_list":["post-238469","post","type-post","status-publish","format-standard","hentry","category-biotech-medical","category-robotics-ai"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/238469","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/users\/427"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=238469"}],"version-history":[{"count":0,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/238469\/revisions"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=238469"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=238469"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=238469"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}