IntroductionPsychotic disorders such schizophrenia and attention-deficit/hyperactivity disorder (ADHD) are neurodevelopmental disorders with social cognitive deficits. Specifically, biased interpretation of social information can result in interpersonal difficulties. Cognitive biases are prevalent in psychosis, but no previous study has investigated whether the type and severity of cognitive biases differ between subjects experiencing first-episode psychosis (FEP) with (FEP-ADHD+) and without ADHD (FEP-ADHD−).MethodsA total of 121 FEP outpatients at the Early Intervention Service of Reus were screened for childhood ADHD through the Diagnostic Interview for ADHD (DIVA). Cognitive biases were assessed by the Cognitive Biases Questionnaire for Psychosis (CBQp). CBQp scores of FEPs groups were compared with those of healthy controls (HCs) with an analysis of covariance. Spearman correlation analysis explored associations between CBQp scores and psychopathology.ResultsThirty-one FEPs met the criteria for childhood ADHD and reported significantly more cognitive bias [median (interquartile range): 47 (38–56)] than FEP-ADHD− [42 (37–48)] and HCs [38 (35.5–43)]. CBQp scores did not differ between FEP-ADHD-and HCs when adjusted for age and sex. After controlling for clinical differences, Intentionalising (F = 20.97; p < 0.001) and Emotional Reasoning biases (F = 4.17; p = 0.04) were more strongly associated with FEP-ADHD+ than FEP-ADHD−. Cognitive biases were significantly correlated with positive psychotic symptoms in both groups but only with depressive symptoms in FEP-ADHD− (r = 0.258; p = 0.03) and with poor functioning in FEP-ADHD+ (r = −0.504; p = 0.003).ConclusionCognitive bias severity increased from HCs to FEP-ADHD-patients to FEP-ADHD+ patients. FEP-ADHD+ patients may be a particularly vulnerable group in which metacognitive targeted interventions are needed.