Ilker Yildirim
Assistant ProfessorCards
About
Research
Publications
2026
Computational models reveal intuitive physics and statistical cues separately contribute to the visual perception of liquids
Zhang Y, Bi W, Miao Y, Yildirim I. Computational models reveal intuitive physics and statistical cues separately contribute to the visual perception of liquids. Cognition 2026, 271: 106469. PMID: 41653826, DOI: 10.1016/j.cognition.2026.106469.Peer-Reviewed Original ResearchConceptsVisual perceptionHigh-level imageStatistical representationPhysics-based representationsComputer visionSimilarity judgmentsComplex intrinsic dynamicsImage statisticsModel architectureComputational basisHuman judgmentRigid objectsPerceptual representationsIntuitive physicsNon-flat surfacesDifferent scenariosQualitatively different scenariosComputational modelDNNStatistical cuesIntrinsic dynamicsVisual experience
2025
A new algorithm of human attention – ERRATUM
Belledonne M, Yildirim I. A new algorithm of human attention – ERRATUM. Behavioral And Brain Sciences 2025, 48: e193. PMID: 41414791, DOI: 10.1017/s0140525x25104524.Peer-Reviewed Original ResearchA new algorithm of human attention
Belledonne M, Yildirim I. A new algorithm of human attention. Behavioral And Brain Sciences 2025, 48: e131. PMID: 41292476, DOI: 10.1017/s0140525x25000196.Peer-Reviewed Original ResearchComputational models reveal that intuitive physics underlies visual processing of soft objects
Bi W, Shah A, Wong K, Scholl B, Yildirim I. Computational models reveal that intuitive physics underlies visual processing of soft objects. Nature Communications 2025, 16: 6303. PMID: 40634304, PMCID: PMC12241314, DOI: 10.1038/s41467-025-61458-x.Peer-Reviewed Original ResearchConceptsDeep neural networksSoft objectsInvariance to changesVisual perceptionPhysics-based simulationsScene contextIntuitive physicsImage statisticsNeural networkMachine visionRigid objectsHuman performanceComputational modelEveryday environmentsRepresentation of clothingProbabilistic representationObject sizeModel of perceptionMatching taskProperties of clothingQualitatively different challengesHuman cognitionObject perceptionVisual processingExisting modelsMultiarea processing in body patches of the primate inferotemporal cortex implements inverse graphics
Yilmaz H, Shah A, Letrou A, Kumar S, Vogels R, Yildirim I. Multiarea processing in body patches of the primate inferotemporal cortex implements inverse graphics. Proceedings Of The National Academy Of Sciences Of The United States Of America 2025, 122: e2420287122. PMID: 40627399, PMCID: PMC12280979, DOI: 10.1073/pnas.2420287122.Peer-Reviewed Original ResearchConceptsInverse graphicsInference networkIT networksGenerative modelPrimate inferotemporal cortexNeural algorithmMacaque IT cortexUnsupervised variantVision modelsImage statisticsVision capabilitiesTransform sensory inputsComputational neuroscienceObject identityGraphicsNetworkIT cortexNeural representationProcessing networkAlgorithmMap imagesInferotemporal cortexRepresentationInferenceTheory of visionAdaptive Computation as a New Mechanism of Dynamic Human Attention
Belledonne M, Butkus E, Scholl B, Yildirim I. Adaptive Computation as a New Mechanism of Dynamic Human Attention. Psychological Review 2025 PMID: 40569736, DOI: 10.1037/rev0000572.Peer-Reviewed Original ResearchAdaptive computationPerceptual computingHuman attentionOn-the-flyAttention dynamicsObject selectionTarget objectHeuristic componentsComputational termsDynamic algorithmDomain-generalAttentional deploymentCase studyTask-relevantIdentical distractorsComputerComputational mechanicsVisual processingVisual attentionDecision outcomesDynamic operationPast modelsAlgorithmDeploymentGeneralized formulationDynamic point-light cloths generate rich percepts beyond biology
Erdogan M, Bi W, Yildirim I, Scholl B. Dynamic point-light cloths generate rich percepts beyond biology. Current Biology 2025, 35: 2967-2973.e3. PMID: 40398413, DOI: 10.1016/j.cub.2025.04.060.Peer-Reviewed Original Research
2024
Response to Goddu et al.: new ways of characterizing and acquiring knowledge
Yildirim I, Paul L. Response to Goddu et al.: new ways of characterizing and acquiring knowledge. Trends In Cognitive Sciences 2024, 28: 965-966. PMID: 39304392, DOI: 10.1016/j.tics.2024.08.004.Peer-Reviewed Original ResearchAre rich percepts from point-light displays specific to biological motion?: A case study of dynamic point-light cloths
Erdogan M, Bi W, Yildirim I, Scholl B. Are rich percepts from point-light displays specific to biological motion?: A case study of dynamic point-light cloths. Journal Of Vision 2024, 24: 1271. DOI: 10.1167/jov.24.10.1271.Peer-Reviewed Original ResearchEfficient Inverse Graphics with Differentiable Generative Models Explains Trial-level Face Discriminations and Robustness of Face Perception to Unusual Viewing Angles
Yilmaz H, Muellner M, Tenenbaum J, Dobs K, Yildirim I. Efficient Inverse Graphics with Differentiable Generative Models Explains Trial-level Face Discriminations and Robustness of Face Perception to Unusual Viewing Angles. Journal Of Vision 2024, 24: 1354. DOI: 10.1167/jov.24.10.1354.Peer-Reviewed Original Research