Abstract: Numerous studies have demonstrated that there is a processing bias existing in the human brain toward self-related information rather than other-related information, such as one’s own face and name. Self-related information, due to its high social/adaptive value, seems to have a preferential access to our attentional resources (the cocktail party effect) and memory system (self-reference effect). Electrophysiological studies confirmed this effect and found that self-relevant information induced enhanced ERP activations. However, in these studies on self, self-related information were adopted in concrete object. Previous event-related potential (ERP) studies showed that phase-locked electrophysiological activities mediate self-relevant possessive pronoun processing relative to non-self-relevant possessive pronoun “his”, self-relevant possessive pronoun “my” elicited a signi?cantly increased P300 amplitude. It means that even if separated from the specific object, abstract self-relevant possessive pronouns also obtained an advantage processing. However, whether non-phase-locked neural oscillations are involved in self-relevant possessive pronoun remains unknown. The aim of this study is to investigate the functional role of non-phase-locked neural oscillations in encoding self-relevant possessive pronoun. The experiment was adopted a three-stimulus oddball paradigm. The experimental materials include four categories, big circle, small circle, and two types of possessive pronouns (“my”, “his”). The big circle was served as standard stimulus (80%), the small circle was served as target stimulus (10%) and two categories of possessive pronouns (“my” (5%), “his” (5%)) distractors. The task of the participants was to observe the stimuli carefully and make behavioral response to the small circle. Twenty healthy students participated in this experiment. The electroencephalogram (EEG) was continuously recorded from scalp electrodes using the 256-channel HydroCel Geodesic Sensor Net (Electrical Geodesics, Inc., Eugene, OR) while subjects were performing the tasks. Time-frequency analysis of the data was conducted using wavelet-based analysis. The maximum amplitude of the mean power of theta (4~8Hz) was extracted from a 0~800ms time-window, for each participant. The extracted power data were then analyzed using repeated measures ANOVAs, with the possessive pronouns (my, his) and the location as within-subject variables. Wavelet analysis result shows that, relative to non-self-relevant possessive pronoun “his”, self-relevant possessive pronoun “my” elicited a signi?cant higher theta-wave (4~8 Hz) power in a large range of scalp. Our EEG results provide evidence showing that, similar to the phase-locked ERPs, non-phase-locked neural activities were also modulated the processing of self-relevant possessive pronouns, and this modulation involves the interplay of multiple cortices. Based on the existing related research, in this study, the large-scale increase in theta power that was found in processing self-relevant possessive pronouns might reflect the attention preference triggered by its intrinsic psychological significance and the encoding and retrieval of abundant information that can be called in “My”.

Key words: self, self-relevant possessive pronoun, theta wave, neural oscillations