Model-based experimental manipulation of probabilistic behavior in interpretable behavioral latent variable models

• Verfasst von: Thome, Janine [VerfasserIn]
• Verfasst von: Pinger, Mathieu [VerfasserIn]
• Verfasst von: Durstewitz, Daniel [VerfasserIn]
• Verfasst von: Sommer, Wolfgang H. [VerfasserIn]
• Verfasst von: Kirsch, Peter [VerfasserIn]
• Verfasst von: Koppe, Georgia [VerfasserIn]
• Titel: Model-based experimental manipulation of probabilistic behavior in interpretable behavioral latent variable models
• Verf.angabe: Janine Thome, Mathieu Pinger, Daniel Durstewitz, Wolfgang H. Sommer, Peter Kirsch and Georgia Koppe
• E-Jahr: 2023
• Jahr: 09 January 2023
• Umfang: 14 S.
• Fussnoten: Gesehen am 28.02.2023
• Titel Quelle: Enthalten in: Frontiers in neuroscience
• Ort Quelle: Lausanne : Frontiers Research Foundation, 2007
• Jahr Quelle: 2023
• Band/Heft Quelle: 16(2023), Artikel-ID 1077735, Seite 1-14
• ISSN Quelle: 1662-453X
• DOI: doi:10.3389/fnins.2022.1077735
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 1837769338

Abstract

IntroductionInterpretable latent variable models that probabilistically link behavioral observations to an underlying latent process have increasingly been used to draw inferences on cognition from observed behavior. The latent process usually connects experimental variables to cognitive computation. While such models provide important insights into the latent processes generating behavior, one important aspect has often been overlooked. They may also be used to generate precise and falsifiable behavioral predictions as a function of the modeled experimental variables. In doing so, they pinpoint how experimental conditions must be designed to elicit desired behavior and generate adaptive experiments.MethodsThese ideas are exemplified on the process of delay discounting (DD). After inferring DD models from behavior on a typical DD task, the models are leveraged to generate a second adaptive DD task. Experimental trials in this task are designed to elicit 9 graded behavioral discounting probabilities across participants. Models are then validated and contrasted to competing models in the field by assessing the ouf-of-sample prediction error.ResultsThe proposed framework induces discounting probabilities on nine levels. In contrast to several alternative models, the applied model exhibits high validity as indicated by a comparably low prediction error. We also report evidence for inter-individual differences with respect to the most suitable models underlying behavior. Finally, we outline how to adapt the proposed method to the investigation of other cognitive processes including reinforcement learning.DiscussionInducing graded behavioral frequencies with the proposed framework may help to highly resolve the underlying cognitive construct and associated neuronal substrates.