It is important to understand the development of depression, and how a family component to treatment affects early adolescent girls’ depression given the association between depression and negative future outcomes. A potential vulnerability to depression is the cognitive triad, which encompasses beliefs about the self, world, and future and is shaped by early learning experiences. Research indicates that the vulnerability originates from parent-child relationships and family messages, which are important in the development of youth cognitive styles. Previous research also indicates that family variables are important factors to consider in the treatment of depression. The purpose of the current study was to expand previous research by examining the roles of perceived family messages and the cognitive triad in the development of depression for early adolescent girls. The study also explored whether parent gender differentiated how family messages affect girls’ cognitions and depressive symptoms. The study evaluated how the addition of a parent training component to a school-based, group-administered CBT intervention affected change in the model’s variables in comparison to group-administered CBT intervention alone and a monitoring control condition. Participants included early adolescent girls diagnosed with depression and caregivers in the parental treatment component. Girls were randomly assigned to a CBT, CBT+PT, or minimal contact control condition. Ratings of girls’ perceptions of family messages, cognitions, and depressive symptoms were obtained at pre-treatment and post-treatment. Results from structural equation modeling indicated significant effects from perceived family messages to girls’ cognitions. Further, girls’ cognitions mediated the relation between perceived family messages and girls’ depressive symptoms at post-treatment for participants within the CBT and monitoring conditions. No significant differences were evident between parent gender and perceived messages. Results indicated that the addition of a parent component to the CBT intervention did not significantly differ from the CBT intervention alone in its effects on the variable relations within the model at post-treatment. Supplemental analyses highlighted parent attendance as a significant factor, with larger effects from the family messages on girls’ cognitions appearing when parents attended majority (six or more) of the eight parent training sessions. Implications, limitations, and areas for further research are discussed. / text