Sensing minimal self in sentences involving the speaker

The role of language in the narrative self is well-known, but does it also affect the minimal self? We investigated whether variations in sentence structure affect the speaker's sense of the minimal self. Previous research has examined how third-person subject expressions influence interpretations of agency, particularly in causal contexts. We examined whether expressing the first-person subject's involvement in causation and perception events influences the speaker's sense of agency and ownership, key components of the minimal self. Participants completed an online experiment using psychological rating scales to evaluate Japanese sentences with varying degrees of speaker involvement, as if they had uttered them. Each sentence varied in whether and how it encoded causation or perception. We analyzed Japanese data since, in addition to perceiver-prominent sentences (e.g., [watashi ga] hoshi o mita, “I saw a star”), Japanese has perceiver-stimulus-prominent sentences (e.g., [watashi ni] hoshi ga mieta, “A star was visible to me” / “I could see a star”), which also foreground the first-person perceiver in a double subject construction. We found that the speaker's sense of agency was significantly lower in sentences where either causation or perception was absent (e.g., hoshi ga deteita, “A star was out”) compared to sentences where either was present. Agency was also significantly higher for perceiver-prominent sentences than for perceiver-stimulus-prominent sentences. Regarding ownership, it was likewise significantly lower when either causation or perception was absent. Whether ownership was significantly higher for perceiver-prominent sentences than for perceiver-stimulus-prominent sentences varied depending on the perceived stimulus. These results suggest that variations in linguistic structure can distinctly impact the senses of agency and ownership. In cognitive linguistics, certain sentence structures are analyzed to reflect how subjectively the speaker is construed. Our findings suggest that differences in agency and ownership provide an empirical basis for this argument, grounded in embodied experience.