Long-term memory encoding of facial information in humans: an EEG and tACS study

In recent years, the investigation of memory formation and retrieval has attracted increasing interest. As oscillatory activity plays a crucial role in neuroplastic processes, episodic memory is to a considerable extent attributable to synaptic changes, synchronization, and neurophysiological alterations through oscillating electric fields. Perception processes are part of episodic memory encoding. Human face perception and encoding arouse particular interest due to their fundamental relevance in social behavior. This study aimed to determine the causal role of brain dynamics in the encoding of facial episodic memory in humans. As recent studies revealed an enhancement in cognitive processes by the entrainment of internal brain oscillations, tACS stepped up as a new method of non-invasive brain stimulation to induce neuroplasticity (Antal et al. 2008; Antal and Herrmann 2016). It is a promising tool to test the role of brain oscillations on episodic memory encoding in humans and the potential for memory improvement. For the entire study, we developed a memory task that includes encoding, a Short-Term Memory Retrieval Part, maintenance, and a Long-Term Memory Retrieval Part. In the longterm face recognition, we assessed both the performance and the choice confidence on the 3-point scale. Two consecutive experiments were performed. For the first experiment (20 participants), we used 128-channel EEG to identify the region of the brain that is exclusively responsible for the long-term face encoding and the frequency of the brain rhythm that is associated with the successful subsequent recognition. Then, we conducted the tACS experiment (25 participants) based on the frequency and spatial data from the EEG experiment. We implemented a double-blinded, randomized, counterbalanced, crossover, and placebo-controlled study design. 20 minutes of 4 Hz-tACS at 3 mA peak-to-peak were applied during the encoding task to the identified right or to the left TPO area for active control. One more session included sham stimulation to one or the other area. The EEG study revealed a significant synchronization of brain oscillations during successful long-term facial memory encoding in the right TPO area at the low theta range (4 Hz). In complete agreement, the placebo-controlled tACS study showed a significant enhancement of long-term memory recognition performance after the low theta-stimulation of the right but not the left TPO area. The results indicate that low theta oscillations in the right TPO area are vital for successful episodic long-term memory encoding of facial stimuli. Secondly, we confirm that active low theta-tACS applied on this area during encoding improves the subsequent memory recognition performance. This supports the concept of lateralization for face processing in the right posterior brain region; moreover it puts forward this area as a crucial neocortical node in communication with the hippocampus for the long-term memory encoding (Pitcher et al. 2011; Geib et al. 2017). The results are in agreement with other studies that revealed a positive correlation between theta power and memory performance (Pahor and Jaušovec 2014; Clouter et al. 2017). However, the present work reveals a causal link between the empowered low theta in the right TPO area and enhanced subsequent long-term memory recognition. In summary, tACS is a highly suitable non-invasive tool to entrain local neocortical low theta activity and enhance long-term memory encoding, which is important in the clinical context for revealing novel therapeutic strategies in prosopagnosia and prosopamnesia.2021-09-1