In this study we test one of the key assumptions of the analysis of wrong answers on the Standard Progressive Matrices test approach — the assumption that “behind the correct answer lies the application of the correct principle for solving particular problem”. This popular alternative approach to psychometric studies of intelligence is based on assumptions that have not been empirically tested. After applying thematic analysis to the transcripts of interview-conversations with 9 and 11 year olds, who solved test tasks out loud, we were able to refute this assumption. We were able to claster identified explanations for the correct answers into several groups, which include both the application of the correct principle, as well as a series of incorrect ones — for example, creating a collection or visually complementing a pattern.