<p>Volunteers can help to reduce social isolation in patients with psychosis. However, little is known about the role of patient attachment style in the relationships established between patients and volunteers. This mixed methods study, using a convergent parallel design, aimed to investigate these relationships, and whether patient attachment style affects them. The sample size for both qualitative and quantitative analysis was 34. The Scale to Assess Therapeutic Relationship (Patient and Volunteer versions) and the Revised Adult Attachment Scale were used to investigate the therapeutic relationships and attachment styles. The Spearman’s Rho test revealed a moderate positive correlation between the ratings of the therapeutic relationship of volunteers and patients (<i>r</i> = 0.538, <i>p</i> = 0.032). No significant influence was found of patient attachment style on the relationship with the volunteers [close attachment (<i>r</i> = 0.39, <i>p</i> = 0.16), dependent attachment (<i>r</i> = 0.12, <i>p</i> = 0.66), and anxious attachment (<i>r</i> = − 0.23, <i>p</i> = 0.4)]. Thematic analysis was conducted on the end of study semi-structured interviews. Seven themes were found: support in the patient-volunteer relationship, trust in the volunteer, impact of the relationship on patients and volunteers, communication between patients and volunteers, attitudes towards the relationship between patients and volunteers, traits of a secure attachment style, and traits of an insecure attachment style. No influence was found between patient attachment style and patient-volunteer relationship. Given the brief intervention, further research is required to investigate these phenomena in greater depth.</p>