Human cognition is highly adaptive and can reason flexibly with new information. A crucial element for this is our working memory: a quick, reliable, and flexible system to maintain information temporarily, as needed, when needed. The goal of this project is to advance the understanding of how working memory is implemented in the human brain. To this end, the main objective is to develop a neural model of human working memory.
During the project, we will follow an intertwined model-experimental approach. First, we will use electroencephalography (EEG) experiments to investigate the neural basis of human working memory. Although it was long thought that working memory relied exclusively on active neural mechanisms that store information through continuous neural spiking, it was recently discovered that activity-silent mechanisms, such as short-term synaptic plasticity, also play an important role. We will experimentally target these two mechanisms, using EEG in combination with machine learning to reveal working memory contents held within. Second, we will use computational spiking-neuron models to explain the results of the experiments and implement the neural mechanisms responsible. These models will also shed light on the functional significance of having two mechanisms of working memory storage and how they are employed: we hypothesize that these exist to balance energy use, speed of information access, and susceptibility to interference.
Application deadline: 18 August 2025, or until filled