<p style="text-align: justify;">Contemporary representations of auditory hallucinations have intensified the assessment of voices as meaningful representations of the stress of interpersonal traumatic experiences. The aim of the study was to verify the relationship between social anxiety, voices, shame, and metacognitive strategies and to examine the features of attention distortion and cognitive activity in a conflict task using social-emotional stimuli and feedback in social anxiety. The total sample consisted of 122 individuals (32 males and 90 females; M<sub>age</sub>=26.11, SD<sub>age</sub>=8.60). The clinical group consisted of 49 persons with subpsychotic (n=28; 7 men and 21 women, M<sub>age</sub>=24.54, SD<sub>age</sub>=5.59) and psychotic symptoms (n=21, 12 men and 9 women, M<sub>age</sub>=28.64, SD<sub>age</sub>=11.14). The nonclinical group consisted of 73 subjects (13 men and 60 women, M<sub>age</sub>=25.58, SD<sub>age</sub>=8.51). In both groups, subjects reliably reported thinking objectification, dissociative phenomena, and voices. A subgroup (n=31, 10 men and 21 women, M<sub>age</sub>=19.87, SD<sub>age</sub>=1.22) was selected to participate in the experiment, divided by the level of social anxiety: high (n=16, 4 men and 12 women, M<sub>age</sub>=19.96, SD<sub>age</sub>=1.14) and low (n=15, 6 men and 9 women, M<sub>age</sub>=19.34, SD<sub>age</sub>=1.05). The following techniques were used: Russian-language versions of the Thought Suppression and Rumination Scale (adapted by Dorosheva, Knyazev, 2017); the Fear of Negative Evaluation Scale (adapted by Grigorieva, Enikolopov, 2016); author's Psychotic Experiences Questionnaire (2023), Social Anxiety and Social Anxiety Disorder Questionnaire (2016) and Self-Focused Attention Questionnaire (2016). After the diagnosis of psychological parameters, an eye-tracking experiment was conducted using a modified flanking Ericksen task. It was found that thought suppression (F<sub>(1;120)</sub>=11.11; p=0.001), rumination (F<sub>(1;120)</sub>=7.43; p=0.01), self-focused attention (F<sub>(1;120)</sub>=3.85; p=0.05), social anxiety (F<sub>(1;120)</sub>=7.30; p=0.01), and shame (F<sub>(1;120)</sub>=21.62; p<0.001) than in the absence of such experiences. We obtained a two-factor solution (each factor explains 30% of the variance) reflecting metacognitive pathways of voice formation: 1) self-focus supports social anxiety, shame, a tendency to suppress thoughts, and the emergence of voices; 2) pronounced thought suppression and rumination associated with shame increase the likelihood of voices, with the detectability of social anxiety attenuated. Significant differences (p<0.001) were found between the high and low social anxiety groups in the parameters of eye-movement and behavioral activity in the eye-tracking experiment (e.g., total test passage time, total number of fixations, response time after a mistaken decision, number of fixations of the off-target stimulus — the central face — and the target stimulus). The conclusion is that further research is needed to clarify the direction of the relationship between psychological parameters and voice experience.</p>
