Ron Chrisley

While the recent special issue of JCS on machine consciousness (Volume 14, Issue 7) was in preparation, a collection of papers on the same topic, entitled Artificial Consciousness and edited by Antonio Chella and Riccardo Manzotti, was published. 1 The editors of the JCS special issue, Ron Chrisley, Robert Clowes and Steve Torrance, thought it would be a timely and productive move to have authors of papers in their collection review the papers in the Chella and Manzotti book, and include these reviews in the special issue of the journal. Eight of the JCS authors (plus Uziel Awret) volunteered to review one or more of the fifteen papers in Artificial Consciousness; these individual reviews were then collected together with a minimal amount of editing to produce a seamless chapter-by-chapter review of the entire book. Because the number and length of contributions to the JCS issue was greater than expected, the collective review of Artificial Con- sciousness had to be omitted, but here at last it is. Each paper’s review is written by a single author, so any comments made may not reflect the opinions of all nine of the joint authors!

Standard, linguistic means of specifying the content of mental states do so by expressing the content in question. Such means fail when it comes to capturing non-conceptual aspects of visual experience, since no linguistic expression can adequately express such content. One alternative is to use depictions: images that either evoke (reproduce in the recipient) or refer to the content of the experience. Practical considerations concerning the generation and integration of such depictions argue in favour of a synthetic approach: the generation of depictions through the use of an embodied, perceiving and acting agent, either virtual or real. This paper takes the first steps in an investigation as to how one might use a robot to specify the non-conceptual content of the visual experience of an (hypothetical) organism that the robot models.

A connectionist system that is capable of learning about the spatial structure of a simple world is used for the purposes of synthetic epistemology: the creation and analysis of artificial systems in order to clarify philosophical issues that arise in the explanation of how agents, both natural and artificial, represent the world. In this case, the issues to be clarified focus on the content of representational states that exist prior to a fully objective understanding of a spatial domain. In particular, the criticisms of (Chrisley, 1993) that were raised in (Holland, 1994) are addressed: how can we determine that a system’s spatial representations are more objective than before? And under what conditions (tasks, training regimes, environments) do such increases in objectivity occur? After analysing the results of experiments that attempt to shed light on these questions, the study concludes by comparing and contrasting this work with related research.

The term \synthetic phenomenology" refers to: 1) any attempt to characterize the phenomenal states possessed, or modeled by, an artefact ; or 2) any attempt to use an artefact to help specify phenomenal states. The notion of synthetic phenomenology is clari¯ed, and distinguished from some related notions. It is argued that much work in machine consciousness would bene¯t from being more cognizant of the need for synthetic phenomenology of the ¯rst type, and of the possible forms it may take. It is then argued that synthetic phenomenology of the second type looks set to resolve some problems confronted by standard, nonsynthetic attempts at characterizing phenomenal states. An example of the second form of synthetic phenomenology is given

Velmans’ paper raises three problems concerning mental causation: (1) How can consciousness affect the physical, given that the physical world appears causally closed? 10 (2) How can one be in conscious control of processes of which one is not consciously aware? (3) Conscious experiences appear to come too late to causally affect the processes to which they most obviously relate. In an appendix Velmans gives his reasons for refusing to resolve these problems through adopting the position (which he labels ‘physicalism’) that ‘consciousness is nothing more than a state of the brain’. The rest of the paper, then, is an attempt to solve these problems without embracing a reductionist physicalism. Velmans’ solution to the first problem is ‘ontological monism combined with epistemological dualism’: First-person and third-person accounts are two different ways of knowing the same facts. This kind of reply is not new; it is, for example, a twist on the position expressed in Davidson (1970). True, there are substantial differences: For one, Davidson reconciles the tension between descriptions of events in mentalistic and physicalist language, not between firstand third-person descriptions of states; for another, Davidson actually provides an argument for his position, although to do so he assumes that there are no psycho-physical (or indeed, psycho-psycho) laws, something which I suspect Velmans would be reluctant to do. Nevertheless, they have in common the idea that the causal efficacy of the mental is not at odds with the causal closure of physics, since a mind-involving causal story is just another way of talking about the same facts that a purely physical causal story talks about. This ‘dual-aspect’ approach is a popular tactic for resolving the mind--body problem, but it has some well-known problems, and it is unfortunate Velmans doesn’t reply to these standard objections. For example, a frequently discussed issue in connection with theories of mental causation is the problem of overdetermination (see, e.g., Unger, 1977; Peacocke, 1979).

Some foundational conceptual issues concerning anticipatory systems are identified and discussed: 1) The doubly temporal nature of anticipation is noted: anticipations are directed toward one time, and exist at another; 2) Anticipatory systems can be open: they can perturb and be perturbed by states external to the system; 3) Anticipation may be facilitated by a system modeling the relation between its own output, its environment, and its future input; 4) Anticipations must be a part of the system whose anticipations they are. Each of these points are made more precise by considering what changes they require to be made to the basic equation characterising anticipatory systems. In addition, some philosophical questions concerning the content of anticipatory representations are considered.

It is suggested that some limitations of current designs for medical AI systems stem from the failure of those designs to address issues of artificial consciousness. Consciousness would appear to play a key role in the expertise, particularly the moral expertise, of human medical agents, including, for example, autonomous weighting of options in diagnosis; planning treatment; use of imaginative creativity to generate courses of action; sensorimotor flexibility and sensitivity; empathetic and morally appropriate responsiveness; and so on. Thus, it is argued, a plausible design constraint for a successful ethical machine medical or care agent is for it to at least model, if not reproduce, relevant aspects of consciousness and associated abilities. In order to provide theoretical grounding for such an enterprise we examine some key philosophical issues that concern the machine modelling of consciousness and ethics, and we show how questions relating to the first research goal are relevant to medical machine ethics. We believe that this will overcome a blanket skepticism concerning the relevance of understanding consciousness, to the design and construction of artificial ethical agents for medical or care contexts. It is thus argued that it would be prudent for designers of MME agents to reflect on issues to do with consciousness and medical expertise; to become more aware of relevant research in the field of machine consciousness ; and to incorporate insights gained from these efforts into their designs.