Causal knowledge is fundamental to learning and instruction. When a student learns, they consolidate causal knowledge about the subject; the student has learned “why” something is the way it is or why an action is taken. Teachers support causal learning by providing causal explanations. Such explanations do more than instruct the student on the correct information or procedure. They also inform the student of why that information or procedure is correct by explicating what information or procedure is wrong. This identification of erroneous counterfactual procedures can allow a student to make the correct inferences necessary to achieve some learning outcome. Currently, this feature of learning is largely unexplored in Intelligent Tutoring Systems. We propose a novel extension of the four-component architecture paradigm that accounts for counterfactual causal explanations by unifying impasse-driven learning, model tracing, constraint-based modelling, and counterfactual causation.