Based on the attachment promotion hypothesis of dreaming, this study explored the wake-to-dream continuity of youths' attachment representations and early maladaptive schemas. A total of 19 children and adolescents (12 boys; M = 11.8 years, SD = 2.1) were administered the Child Attachment Interview and completed the self-reported Schema Inventory for Children. They were provided a digital audio recorder to report their dreams for 14 consecutive days. Data from 95 dreams (M = 5 per participant, SD = 2.6) were collected. Dream coding systems were developed to assess attachment-and schema-related content. Bootstrapped correlations were conducted between waking scores and the highest dream-derived scores. In addition, a cluster analysis of dreamswith a subsequent inclusion of the waking scores-was conducted for each participant. The results showed that security-related waking scores were linked to insecurity-related dreaming scores (dismissingness). Conversely, insecurity-related waking scores (dismissingness) were negatively linked to security-related dreaming scores. Security-related waking scores accounted significantly for the merging of the waking attachment profile to a cluster of dreams. No wake-to-dream association was found for attachment-related schemas (Disconnection and Rejection domain). Because dreams depict more insecure than secure attachment-related content, they may function to rehearse different attachment strategies. However, attachment security showed wake-to-dream continuity through cluster analysis, which might be a way to consolidate, in addition to a way to test or improve, attachment representations in childhood and adolescence. Implications for the hypothesized dream functions of attachment promotion, emotion regulation, and threat simulation are discussed.