Oszillatorische und fraktale Dynamiken: Wie latente Stimuluseigenschaften das menschliche Arbeitsgedächtnis formen

01.08.2023 bis 31.03.2025

Working memory (WM) is orchestrated in the human prefrontal cortex (PFC). Depending on the level of observation, several physiological underpinnings of WM have been proposed. These mechanisms range from irregular, persistent neuronal spiking on the cellular level to synchronized neural oscillations in field potentials. So far, human research focused on neural oscillations that constitute temporal reference frames where information can be maintained and integrated. However, recent evidence suggests that irregular (also termed fractal) brain activity, which follows a 1/f power law and has been shown to dominate the background activity in electrophysiological recordings also supports WM-processing. To date, it remains unclear how oscillatory and fractal neural activity operate in concert to coordinate WM. Interestingly, stimuli that are frequently used to probe WM exhibit latent but striking parallels to neural activity when transformed into the frequency domain: Gabor gratings exhibit oscillatory features whereas natural stimuli exhibit fractal features. These (non-) oscillatory stimulus features are accurately tracked by their corresponding neural responses in sensory areas. This process is thought to promote efficient neural processing. However, if and how these stimulus features affect WM processing in higher-order association areas such as the PFC remains undetermined. This project has two objectives: first, elucidate the distinct contribution of oscillatory and fractal brain dynamics during WM processing, and second, unravel their functional interactions with stimulus features in higher-order cortical areas. I plan to directly contrast highly controlled, artificial stimuli with oscillatory and fractal features during WM processing. I predict that oscillatory and fractal stimuli promote their respective neural activity during WM processes. I further hypothesize that the PFC, like sensory areas, exploits latent stimulus features for efficient WM-processing. To formally test these hypotheses, I will conduct magnetoencephalography in healthy participants and intracranial electroencephalography in pre-surgical epilepsy patients in combination with behavioral testing. I will utilize state-of-the-art spectral analyses for neural and stimulus data to disentangle oscillatory and fractal characteristics. Additionally, I will generate a highly parameterized fractal stimulus set to precisely gauge the fractal degree of these stimuli. The outcome of this project will advance our understanding of how exogenous and endogenous oscillatory and fractal dynamics jointly shape human WM.