Emergence of prefrontal neuron maturation properties by training recurrent neural networks in cognitive tasks

Yichen Henry Liu; Junda Zhu; Christos Constantinidis; Xin Zhou

doi:10.1016/j.isci.2021.103178

Emergence of prefrontal neuron maturation properties by training recurrent neural networks in cognitive tasks

iScience. 2021 Sep 27;24(10):103178. doi: 10.1016/j.isci.2021.103178. eCollection 2021 Oct 22.

Authors

Yichen Henry Liu¹, Junda Zhu², Christos Constantinidis^{2

3

4}, Xin Zhou^{1

3

5}

Affiliations

¹ Department of Computer Science, Vanderbilt University, Nashville, TN 37235, USA.
² Neuroscience Program, Vanderbilt University, Nashville, TN 37235, USA.
³ Department of Biomedical Engineering, Vanderbilt University, Nashville, TN 37235, USA.
⁴ Department of Ophthalmology and Visual Sciences, Vanderbilt University Medical Center, Nashville, TN 37232, USA.
⁵ Data Science Institute, Vanderbilt University, Nashville, TN 37235, USA.

Abstract

Working memory and response inhibition are functions that mature relatively late in life, after adolescence, paralleling the maturation of the prefrontal cortex. The link between behavioral and neural maturation is not obvious, however, making it challenging to understand how neural activity underlies the maturation of cognitive function. To gain insights into the nature of observed changes in prefrontal activity between adolescence and adulthood, we investigated the progressive changes in unit activity of recurrent neural networks as they were trained to perform working memory and response inhibition tasks. These included increased delay period activity during working memory tasks and increased activation in antisaccade tasks. These findings reveal universal properties underlying the neuronal computations behind cognitive tasks and explicate the nature of changes that occur as the result of developmental maturation.

Keywords: Cognitive neuroscience; Neural networks; Neuroscience.

Abstract

Grants and funding