Neural Correlates of Self Awareness

Without any doubt the human ability of self awareness is puzzling. Up to date, various approaches have been realized to augment the understanding of this fascinating phenomenon. The goal of the present project is to contribute to this active field of research by focusing on a quantitative descrip-tion of the temporal properties of brain activities related to self awareness. Experimental designs which control the involvement of neural systems related to self awareness are selected. For this purpose, visually/acoustically presented nouns in combination with different pronouns such as "a" or "my" are used to selectively manipulate the engagement of neural activity related to self awareness. The involvement of self awareness is not only manipulated by the selected pronouns but also by the level of verbal information processing. Shallow and deep processing levels are experimentally controlled and recognition performance of the above mentioned conditions is investigated. Within this approach the engagement of neural activity related to self awareness can be specified in relation to different levels of word processing. To reliably describe the temporal properties of brain activity related to self awareness the magnetoencephalography (MEG) representing a modern brain imaging method which provides an excellent temporal resolution and a respectable spatial resolution is chosen.

Without doubt the human ability of self-awareness is one of the most fascinating and puzzling phenomena in current brain research. This ability is based on physilogical processes in the human brain, which are up to now, basically not understood. The aim of the project was to contribute to this active field of research by clarifying to what extend it is possible to experimentally quantify and localize (in time and space) those neurophysilogical activity patterns in the brain, which are related to `self-awareness`. To this end we used the three actual brain imaging techniques based on electroencephalography (EEG), magnetoencephalography (MEG) and functional magnetresonance imaging (fMRI). The experimental design of this pilot project was such that the neural processing of the pronouns (my, a, his) preceeding a list of nouns was recorded with these three techniques. The depth of word processing can be controlled experimentally. With this setup it should be possible to detect differences in neural activity in association with different pronouns, which could lead to a spatial and temporal localization of the relevant neurophysiological processes, which are related to `self-awareness`. This requires the hypothesis that the notion of self-awareness is non-trivially coupled to language and particular components of it. Since the expected effects are very small, the high temporal of EEG and MEG, as well as the high spatial resolution of the fMRI were prerequisites to the study. That the effects should be in principle detectable we could show in a preceeding study with stutterers, where differences in word processing was detected in stutterers when compared to non-stutterers. The most relevant findings of our project is that we could indeed detect differences in word processing within our experimental setup. Basically we were able to distinguish two effects taking place at two different time periods: The first effect in a time window between 200 and 400 milliseconds after word presentation, the second in a window of 500 to 800 milliseconds. The first effect is found at the occipito-parietal sensors, while the second is predominatly found in the left frontal regions. In both cases there seem to be pronoun-dependent differences in brain activity, regardless of word processing depth. We gathered evidence that within the early processing stage a discrimination between person-related / non-person-related could take place, whereas in the second a discrimination between self / other seems to take place. Our work shows - in agreement with recent literature - that those brain regions which are responsible for those neurophysilogical processes which discriminate between self and non-self, should be indeed detectable, and that different tasks are performed distinct time regions. Since our results are based on three independent brain imaging techniques, we are in the position to evaluate the pros and cons of each method in the context of our experimental paradigm. The datapool will be further analyzed to gain more detailed insights to the question of spatio-temporal localization of word-based pronoun-dependent brain activity.