The present paper is the result of a debate at the 2nd Meeting of Hispa-noamerican Social Psychology held in Madrid in September 1981. Immediate ly following debates concerning the border lines between Cognitive Maps and Behaviour Maps, we set ourselves the task of organizing an introduc-tory seminar on the subject at the Faculty of Arts and Sciences of Educa tion of the University of Barcelona. This paper then, may be read as a reflection upon that seminar. We attempt, basically, to study the theore tical and methodological conditions in which comparison between Cogniti-ve Maps and Behaviour Maps would make sense, but in order to achieve this we first analyne them separately. As we feel that the question of maps has not been too outlined, we have not-dwelt on a system of definitions from the start. What Cognitive Maps and Behaviour Maps may be should be molded in the course of the following considerations. In spite of this, our stance at the start of the seminar and at the begining of this article may be outlined as follows: a) Cogni-tive Maps may be taken as simulations of the cognitive organization of space, which is a product of the subject's activity, or otherwise as the organization itself(for example Stea and Downs, 1977 p.6); as you will see this double meaning heralds the difficulties we shall encounter further on. b) Behaviour Maps are representations on plans (and other projections) or tables of observational data concerning the spatial behaviour of the people seen fundamentally as an interaction with the physical world (for example, Proshansky, Ittelson and Rivlin, 1978 pp 845-846). We trust that it will become patent that these conceptions, apparently wise, hide serious difficulties under a barely critical perspective.