We conducted four lab and online experiments, and a field experiment to test our hypotheses and to rule out an alternative explanation. The design of each of the five experiments consisted of presenting participants with two sequential choices. In our experiments we manipulated the expected visibility of the product involved in the first decision, and evaluating whether the identity type (social or personal) emphasized in the second choice matched the identity type emphasized in the first choice. Emphasizing the same identity in both choices represents consistency in identity-related sequential decisions. Experiments were conducted both in public and in private settings, enabling us to confirm that our results were robust across different levels of decision publicity.