Understanding Advances in Computational Cognitive Neuroscience Over the Past Ten Years: A Scoping Review

Computational Cognitive Neuroscience (CCN) is an interdisciplinary field that uses computationalmodels to understand how the brain gives rise to cognition. Although the field has grown significantlyover the past 10 years, there is no comprehensive review, carried out systematically, of the majoradvances achieved during this period. Therefore, this review aims to map CCN advances over the lastdecade by analyzing the most frequently used computational models, investigated cognitive domains,scientific production, and computational tools used, as well as determining the predominant emphasisof the studies, whether theoretical, applied, or both. This study consists of a scoping review conductedin accordance with the methodology proposed by the Joanna Briggs Institute (JBI). A search wasperformed in the PubMed, Embase, and IEEE Xplore databases. This resulted in the inclusion of 75articles, the analysis of which revealed that recent advances in the field are predominantly theoretical.A notable concentration of studies was observed in the areas of memory, learning, language andcommunication, and decision-making, which have consolidated themselves as the primary researchfronts. Among the selected studies, the two main models were those that used Recurrent NeuralNetworks (RNNs) and those based on Bayesian models. Temporal analysis confirmed an accelerationin the publication rate, peaking in 2021. It is concluded that CCN has advanced quickly over thelast decade, particularly in a theoretical way, yet it still faces challenges regarding neurobiologicalplausibility and model integration, difficulties in study generalization, and the need for further researchwith practical applicability.