Children's manipulations and their answers to the test questions were further evaluated in pairs to determine whether children had shown an understanding of beliefs. If a child's manipulation had disclosed the treasure's location, the test question would be a true-belief question which asked about a belief that corresponded to reality (i.e., where the treasure really was) and answering it correctly would suggest an understanding of true beliefs. On the other hand, if a child's manipulation had not disclosed the treasure's location and s/he even had laid false trails, the test question would then be a false-belief question which asked about a belief that conflicted with reality and answering it correctly would suggest an understanding of false beliefs. Thus, if children passed a true-belief question (most likely in the cooperative trial) and a false-belief questions (most likely in the competitive trial), they were regarded as having demonstrated an understanding of both true and false beliefs and were scored as a ‘2.’ If children passed none of these questions, they were regarded as not having demonstrated any belief understanding and were scored as ‘0.’ If children passed either one type of these questions, they were regarded as having demonstrated the corresponding type of belief understanding and were scored as a ‘1.’