{"slug": "an-irt-shaped-practice-score-is-not-an-iq-test", "title": "An IRT-shaped practice score is not an IQ test", "summary": "IntelligenceMax uses an item-response-theory-shaped formula to update a learner's practice score after each answer, displayed on the IQ scale. However, the developer warns that the score is not a normed IQ test because the AI-generated item parameters are not calibrated from a norming sample, limiting the estimate to an adaptive practice tool rather than a clinical measure.", "body_md": "IntelligenceMax updates a learner estimate after every answer and displays it on the familiar IQ scale. The arithmetic resembles item response theory. That resemblance is useful, but it is easy to ask the formula to carry more meaning than it can bear.\n\nA formula can be internally consistent while its inputs remain uncertain. Here, the distinction begins with the item parameters.\n\nFor ability θ, item difficulty b, and discrimination a, the model assumes a logistic P(correct). The first estimate has a normal prior centred at 100 with SD 15. After an answer, the implementation takes one capped local update using score and Fisher information. Per-answer change is capped at 1.5 points.\n\nThe calculation is the easy part. The inputs are where the uncertainty lives.\n\nThe AI that generates each question also assigns difficulty and discrimination. Those values are checked for shape and range, but they are not fitted from a norming sample. Calling the update IRT-style describes the logistic form. It does not establish that 115 means the same thing as a score of 115 on a validated instrument.\n\nWithin those limits, the estimate can support an adaptive practice loop. It cannot establish a clinical Full Scale IQ, a permanent change in general intelligence, transfer to school or work, or treatment of a cognitive condition.\n\nReviews of commercial brain training find the strongest evidence on trained tasks and close relatives. Broad gains are less convincing against active controls (Simons et al., 2016; Melby-Lervåg et al., 2016).\n\nUntil empirical calibration exists:\n\nThis is a logistic, IRT-shaped practice estimate that uses model-assigned item parameters. It is not a normed IQ score.\n\nPublic notes: [https://intelligencemax.ai/science](https://intelligencemax.ai/science)\n\nEvidence map: [https://intelligencemax.ai/guide](https://intelligencemax.ai/guide)\n\nOSF: [https://osf.io/kja9b/](https://osf.io/kja9b/)\n\nDisclosure: I built IntelligenceMax. The technical question I most want readers to challenge is whether useful empirical calibration is possible for generated, mostly one-off items without quietly turning the system into a fixed bank.", "url": "https://wpnews.pro/news/an-irt-shaped-practice-score-is-not-an-iq-test", "canonical_source": "https://dev.to/connerlambden/an-irt-shaped-practice-score-is-not-an-iq-test-42c8", "published_at": "2026-07-14 22:10:27+00:00", "updated_at": "2026-07-14 22:57:31.056627+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "ai-products", "ai-ethics"], "entities": ["IntelligenceMax", "Simons et al.", "Melby-Lervåg et al."], "alternates": {"html": "https://wpnews.pro/news/an-irt-shaped-practice-score-is-not-an-iq-test", "markdown": "https://wpnews.pro/news/an-irt-shaped-practice-score-is-not-an-iq-test.md", "text": "https://wpnews.pro/news/an-irt-shaped-practice-score-is-not-an-iq-test.txt", "jsonld": "https://wpnews.pro/news/an-irt-shaped-practice-score-is-not-an-iq-test.jsonld"}}