One objection to the preference-satisfaction theory of well-being is that the theory cannot make adequate sense of interpersonal comparisons of well-being. A tradition dating back to at least Harsanyi attempts to solve this problem by means of appeal to so-called 'extended preferences', that is, roughly, preferences over situations whose description includes one’s ordinary (non-extended) preferences. This paper examines the prospects for defending the preference-satisfaction theory via this ‘extended preferences’ program. We argue that making conceptual sense of extended preferences is less problematic than many authors have supposed, but that even so extended preferences do not present an attractive reply to the objection on behalf of the preference- satisfaction theory of well-being. We first present a comparatively superficial challenge to the program, related to Arrow’s celebrated impossibility theorem. We suggest two ways in which the extended-preference theorist might avoid this particular problem: by appeal to aggregation rules (for extended preferences) that violate the Independence of Irrelevant Alternatives condition, and by appeal to a rationality constraint on the relation between preferences and (ideal-preference-determined) betterness facts. Our main challenge to the extended preferences program concerns whether this way of making sense of interpersonal comparisons of well-being is consistent with the core motivations for the preference-satisfaction theory of well being. We argue that it is not.