SU107. Disrupted Continuity of Subjective Time in the Milliseconds Range in the Self-disturbances of Schizophrenia: Convergence of Experimental, Phenomenological, and Predictive Coding Accounts

Background: The impression of time continuity is a pervasive and given property of our subjective life. However, it appears to be compromised in schizophrenia patients who experience what has been labeled “self-disturbances” (Ichstörungen). Both experimental and phenomenological approaches suggest that gaps in continuity of self-experience in schizophrenia patients reflect a disruption of temporal processing on different time scales. We hypothesized that the disruption of time continuity in schizophrenia patients involved an impairment in being able to make predictions about the immediate future, which includes timescales too brief to be experienced consciously. Methods: We applied phenomenological analysis to the experimental findings of a simultaneity/asynchrony implicit judgment task administered to both schizophrenia patients and healthy controls. Subjects had to decide whether the presentation of 2 squares was simultaneous or asynchronous by pressing on the left (simultaneous) or right (asynchronous). Results: When stimulus asynchronies were too brief (<20 ms) to be detected consciously, response biases (Simon effect) were nevertheless observed in the healthy controls by pressing to the side of the second occurring square, whereas patients consistently pressed to the side of the first square. Phenomenological analysis indicated that because the asynchrony is too brief to be experienced consciously, its place in succession is nevertheless coded in terms of a passing retention, which thus accounts for the differential Simon effect in the patients and healthy controls. That is sensory events are processed at the ms level in terms of the retained information even if this information does not reach consciousness. Surprisingly, the unconscious, automatic processing of subjective time is more accurate than conscious processing in both the patients and controls. Experimentally, the patients had difficulties to predict and follow events automatically, at the ms level, consistent with an impairment in predictive coding. These findings were supported by the phenomenological analysis. The responses of the schizophrenia patients suggested an impoverishment in coding the successive information (retentions), which normally informs predictive protentions about what is to happen next even at very brief time scales. These observations are also concordant with hierarchical predictive coding accounts of brain function ie, processing is more local or granular, with smaller timescales lower in the hierarchy. Conclusion: Self-disturbances of schizophrenia reflect disruption of lower nonconscious levels of temporal processing at very small timescales. Whereas the experimental approach highlights prediction impairments, phenomenological analysis provides testable hypotheses regarding the interaction between retention (retained information) and protention (prediction) mechanisms.