Therapist expertise is associated with the use of complex knowledge structures and metacognitive processes. A cross sectional ex-post facto design assessed differences in structural knowledge and metacognitive processes between lay helpers, advanced students, and senior professional therapists. A card sorting task involving 19 therapist intentions was used to assess the following structural knowledge indicators: minutes to complete a card sort, number of card sort categories, and card sort score. Metacognitive processes were assessed using an adaptation of the Metacognitive Awareness Inventory and the Self-reflection subscale of the Self-Reflection and Insight subscales. An inverse U shaped relationship was found in where compared to lay helpers and senior professional therapists; advanced student's had higher card sort scores, indicative of greater consistency with a sample of experienced therapists. Compared to lay helpers and advanced students, senior professional therapists used significantly more time to sort therapist intentions and sorted intentions into a greater number of categories. Relative to metacognitive process, advanced students and senior professional therapists reported significantly greater knowledge of cognition than lay helpers. Also, advanced students also reported greater self-reflection than both lay helpers and senior professional therapists. Discriminant analysis assessed the potential for a linear combination of structural knowledge indicators and metacognitive processes to differentiate participants by level of therapist development. Self-reflection and card sort scores discriminated advanced students from senior professionals, whereas knowledge of cognition and minutes to complete the card sort discriminated experienced professionals from lay helpers. Multidimensional scaling analysis was used to assess the optimal structural configuration of the pooled card sort data and yielded a 4 dimensional solution of the 19 therapist intentions. Results were consistent with Skovholt and Ronnestad's (1992) model of therapist professional development. Results also supported the attenuating effect of ill defined problems on problem solving ability of highly experienced individuals in their respective domain. The study concludes with implications for training, therapy, and research.