Abstract

In the present talk, we will introduce a methodological approach, incorporating multivariate pattern analysis (MVPA) of EEG and eye-tracking techniques. In or previous research, we have successfully utilized this approach to study memory retrieval and mnemonic interference control. Here, we propose how these techniques could be adapted to investigate the switching between language systems in connection with related memory and cognitive control systems.

Our methodological approach utilizes techniques with high temporal resolution. This allows us to capture the temporal dynamics of content-specific neural representations and connect them to distinct cognitive activities, such as memory retrieval and language decoding. Implementing machine learning techniques, MVPA employs classifiers trained to decode specific mental states from distributed patterns of brain activity. These classifiers can then reveal the content of specific mental representations during memory formation or retrieval. For instance, if the visual content of a retrieved memory trace is a human face or an inanimate object. While MVPA has primarily been utilized on MR-data over longer time periods, recent approaches leverage high-temporal resolution EEG to track the dynamic activation of mental representations over time. Members of our research team have been pioneering such techniques in studies of episodic memory and cognitive control. Here, we will present a proposal for applying our approach to decode specific language systems in real-time (e.g., L1 vs L2), and how capturing such dynamics could address fundamental questions concerning how the brain's memory and cognitive control systems interact with the switching between language systems.