Arthur B. Markman (1998) Content and Form are not the Same. Psycoloquy: 9(63) Representation Mediation (7)

Volume: 9 (next, prev) Issue: 63 (next, prev) Article: 7 (next prev first) Alternate versions: ASCII Summary
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 9(63): Content and Form are not the Same

Reply to Terrier on Representation-Mediation

Arthur B. Markman
Department of Psychology
University of Texas
Austin, TX 78712

Eric Dietrich
PACCS Program in Philosophy
Binghamton University
Binghamton, NY


Terrier (1998) suggests that understanding the form of representations will necessarily lead us to an understanding of content. We agree that the questions of form and content are related, and that understanding the form is likely to lead to an understanding of content. Nonetheless, we believe that the issues are separate.


compositionality, computation, connectionism, discrete states, dynamic Systems, explanation, information, meaning, mediating states, representation, rules, semantic Content symbols
1. Terrier (1998) points out that we were careful in our target article (Markman & Dietrich, 1998) to distinguish between the content and form of mediating states. Our definition provides a set of conditions that define something as a mediating state, but it does not describe the content of these states. Terrier suggests that once the form of a mediating state is known, the question of content will necessarily have been answered, and thus they need not be separated.

2. We believe there are two distinct issues that need to be dealt with in understanding the mediating states of a cognitive system. The one that is the focus of our target article is the form of representations. A second is how the mediating states come to be related to other states, and how they remain synchronized to states outside the system. A critical part of our definition of a mediating state is that there is some flow of information between states external and internal to the system. Thus, a theory of representation must ultimately define the way the internal and external states are related. We take the content of a mediating state to be this relation that keeps internal states in registration with external states.

3. There is an appealing insight in Terrier's commentary. An important part of understanding the form of the mediating states used by a given cognitive system will involve learning about how those states get their content. Where we disagree with Terrier is only that we do not claim that understanding the form of a mediating state used by a cognitive system will necessarily lead to a complete understanding of how it got its content. Rather, we leave open the possibility that questions will remain about the content of mediating states even after their form has been established.

4. Part of the reason for keeping these issues separate in our target article is that most cognitive models do not use true mediating states. Instead, they use empty mediating states that are not connected to states external to the system (Bickhard & Terveen, 1995, also make this point). Thus, many debates over representation are based on insights drawn from systems that do not actually represent. These debates focus on what can and cannot be done with particular forms of representation, generally without consideration of their content. Because our target article was designed to provide a set of terms to advance a debate about representation, we focused on the form of representation rather than on the content.


Bickhard, M. H., & Terveen, L. (1995) Foundational issues in artificial intelligence and cognitive science: Impasse and solution. Amsterdam: Elsevier.

Markman, A.B. & Dietrich, E. (1998) In Defense of Representation as Mediation. PSYCOLOQUY 9(48)

Terrier, P. (1998). Why is the question of content still open? PSYCOLOQUY 9(58)

Volume: 9 (next, prev) Issue: 63 (next, prev) Article: 7 (next prev first) Alternate versions: ASCII Summary