2017/04/28 by Sebastian Weichwald, Weichwald, Sebastian, Moritz Grosse‐Wentrup +1
Computer Science · Neuroscience · #Applications (stat.AP) #Cognitive Science and Education Research #Computational Drug Discovery Methods #FOS: Biological sciences #FOS: Computer and information sciences #Functional Brain Connectivity Studies #Methodology (stat.ME) #Neural and Behavioral Psychology Studies #Neural dynamics and brain function #Neurons and Cognition (q-bio.NC)
paper · pdf · doi:10.48550/arxiv.1704.08851
openalex publication_date 2017/04/28 · openalex created_date 2022/09/11 · openalex updated_date 2026/07/28
There are two major questions that neuroimaging studies attempt to answer:\nFirst, how are sensory stimuli represented in the brain (which we term the\nstimulus-based setting)? And, second, how does the brain generate cognition\n(termed the response-based setting)? There has been a lively debate in the\nneuroimaging community whether encoding and decoding models can provide\ninsights into these questions. In this commentary, we construct two simple and\nanalytically tractable examples to demonstrate that while an encoding model\nanalysis helps with the former, neither model is appropriate to satisfactorily\nanswer the latter question. Consequently, we argue that if we want to\nunderstand how the brain generates cognition, we need to move beyond the\nencoding versus decoding dichotomy and instead discuss and develop tools that\nare specifically tailored to our endeavour.\n