**Acknowledgements**

mindreading deficits in delusional-prone individuals. Following our framework on social cognition and consciousness, it is likely that unconscious interpretative contents from the higher-level structures are not properly broadcasted to other systems via global workspace in order to get proper monitoring and evaluations from the low-level structures. Moreover, because the information contents remain unconscious, correction of faulty beliefs and inter-

Interestingly, disturbances in conscious access presented in schizophrenia seem to be confirmed by studies on metacognitive and mindreading malfunctioning by Lysaker and colleagues [64]. Researchers have investigated the impairments within self-generated personal narratives in terms of perceiving one's own state and mental states of others in adults with schizophrenia spectrum disorders. The researchers by using the MAS measure (MAS measure; [55]) have identified three groups of patients: the first one with impaired metacognition ("minimal reflectivity") and poor mindreading, the second group with intact basic self-reflectivity and poor mindreading, and the third group characterized with intact self-reflection and mindreading ability in terms of attributing thoughts and emotions to other people. It turned out that individuals with impaired metacognition and mindreading facilities performed worst in recognizing negative affective cues in others' faces and voices in the video-typed material. According to Dimaggio and colleagues [8], these findings could fit into the concept of the simulation theory (see above) proposed by Gallese and Goldman [65]. Thus, these findings can support the idea of priority of having direct access to one's own internal states to further involve simulation and inferences to interpret and understand others' mental states. Since the study by Lysaker and colleagues [64] was of the correlation nature, alternative interpretations cannot be ruled out. In particular, the study showed that performance results in emotion recognition task were positively correlated with general metacognitive functions (*r* = 0.44) assessed with the "Understanding one's own mind" subscale and to some extent with mindreading capacity (*r* = 0.26) as indicated by scores on the "Decentration" dimension (the ability to perceive others as having their own emotions, thoughts, and perspectives) [64]. These results may have also implication to the idea of dual-process model of social cognition and global workspace. Clearly, better performance in recognizing affective cues in other's faces and voices was linked with activation of both higher- and lower-level structures. Subsequently, the MAS measure "Understanding one's own mind" subscale including also metacognitive function of integration could indicate that low-level metacognitive functions in

emotion recognition were supported by conscious access to some extent.

In our opinion, it becomes clear that efficient interactions between metacognition and mindreading capacities should be supported by access consciousness. The GWT framework by Baars [33, 38] indicates that consciousness can mediate the interaction between metacognition and mindreading subsystems by managing the access to interpretative contents from both mental faculties. Since the GWT assumes the interaction of unconscious and conscious processes, it becomes crucial how and under what conditions interpretative data of a mentalizing

pretations is not possible.

114 Prefrontal Cortex

**8. Conclusions**

This research has been supported by the National Science Centre (Poland) and funded under the award number 2014/15/B/HS6/03834 to RS.
