This paper contrasts two enactive theories of visual experience: the sensorimotor theory (O’Regan and Noë, Behav Brain Sci 24(5):939–1031, 2001; Noë and O’Regan, Vision and mind, 2002; Noë, Action in perception, 2004) and Susan Hurley’s (Consciousness in action, 1998, Synthese 129:3–40, 2001) theory of active perception. We criticise the sensorimotor theory for its commitment to a distinction between mere sensorimotor behaviour and cognition. This is a distinction that is firmly rejected by Hurley. Hurley argues that personal level cognitive abilities emerge out of a complex dynamic feedback system at the subpersonal level. Moreover reflection on the role of eye movements in visual perception establishes a further sense in which a distinction between sensorimotor behaviour and cognition cannot be sustained. The sensorimotor theory has recently come under critical fire (see e.g. Block, J Philos CII(5):259–272, 2005; Prinz, Psyche, 12(1):1–19, 2006; Aizawa, J Philos CIV(1), 2007) for mistaking a merely causal contribution of action to perception for a constitutive contribution. We further argue that the sensorimotor theory is particularly vulnerable to this objection in a way that Hurley’s active perception theory is not. This presents an additional reason for preferring Hurley’s theory as providing a conceptual framework for the enactive programme.

A novel way to model an agent interacting with an environment is introduced, called an Enactive Markov Decision Process (EMDP). An EMDP keeps perception and action embedded within sensorimotor schemes rather than dissociated, in compliance with theories of embodied cognition. Rather than seeking a goal associated with a reward, as in reinforcement learning, an EMDP agent learns to master the sensorimotor contingencies offered by its coupling with the environment. In doing so, the agent exhibits a form of intrinsic motivation related to the autotelic principle (Steels), and a value system attached to interactions called “interactional motivation.” This modeling approach allows the design of agents capable of autonomous self-programming, which provides rudimentary constitutive autonomy – a property that theoreticians of enaction consider necessary for autonomous sense-making (e.g., Froese & Ziemke). A cognitive architecture is presented that allows the agent to discover, memorize, and exploit spatio-sequential regularities of interaction, called Enactive Cognitive Architecture (ECA). In our experiments, behavioral analysis shows that ECA agents develop active perception and begin to construct their own ontological perspective on the environment. Relevance: This publication relates to constructivism by the fact that the agent learns from input data that does not convey ontological information about the environment. That is, the agent learns by actively experiencing its environment through interaction, as opposed to learning by registering observations directly characterizing the environment. This publication also relates to enactivism by the fact that the agent engages in self-programming through its experience from interacting with the environment, rather than executing pre-programmed behaviors.

Excerpt: In this chapter, I argue for the position that active perception is actually best accounted for by a representational theory of perception. Along the way, this will require a relatively novel conception of what to count as representations. In particular, I flesh out a novel account of action-oriented representations: representations that include in their contents commands for certain behaviors.

In this paper, I want to focus on the claim, prominently made by sensorimotor theorists, that perception is something we do. I will argue that understanding perceiving as a bodily doing allows for a strong non-dualistic position on the relation between experience and objective physical events, one which provides insight into why such relation seems problematic while at the same time providing means to relieve the tension. Next I will show how the claim that perception is something we do does not stand in opposition to, and is not refuted by, the fact that we often have perceptual experience without moving. In arguing that cases of motionless perception and perception-like experience are still doings it will be pointed out that the same interactive regularities which are engaged in in active perception still apply to them. Explaining how past interactive regularities can influence current perception or perception-like experience in a way which remains true to the idea that perception is a doing, so I will argue, can be done by invoking the past – the past itself, however, not its representation. The resulting historical, non-representational sensorimotor approach can join forces with Gibsonian ecological psychology – provided that such is also understood along lines that don’t invoke externalist remnants of contents.

Excerpt: Mossio and Taraborelli’s purpose was to revitalize the different families of ecological approaches as proper scientific the- ories by distinguishing the Gibsonian and sensorimotor sub-currents with their different notions of active perception, and then showing how each approach has its scientific merits. We suggest that their analysis is correct if the original enactive approaches such as Varela (1996), Thompson and Varela (2001), and Maturana (2002) are excluded: such neurophenome- nological approaches have no scientifically verifiable link with the external world and should not be classified with senso- rimotor approaches.

This theoretical contribution aims at shedding light on the use of the notion of ‘enactive’ in research on direct perception and on its contribution to the understanding of skill acquisition. While its founders Varela et al. (1991) in the work ‘The Embodied Mind’ have stressed the specificity of the enactive approach and have insisted on the difference between the enactive and the ecological approach, ecological research tends to use the term ‘enactive’ as a general umbrella term to indicate intrinsically active perception. This contribution makes the point that the specificity of the notion of ‘enactive’ cannot be neglected and that this may have some relevant consequences for the understanding and investigation of skill acquisition.

The strategies of action employed by a human subject in order to perceive simple 2-D forms on the basis of tactile sensory feedback have been modelled by an explicit computer algorithm. The modelling process has been constrained and informed by the capacity of human subjects both to consciously describe their own strategies, and to apply explicit strategies; thus, the strategies effectively employed by the human subject have been influenced by the modelling process itself. On this basis, good qualitative and semi-quantitative agreement has been achieved between the trajectories produced by a human subject, and the traces produced by a computer algorithm. The advantage of this “reciprocal modelling” option, besides facilitating agreement between the algorithm and the empirically observed trajectories, is that the theoretical model provides an explanation, and not just a description, of the active perception of the human subject.

Key words: active perception, computer modelling, sensory-motor invariants, visual tactile sensory substitution.

One major criticism of direct or active perception (and other forms of embodied action) from the perspective of cognitive psycology is that, according to common sense, there are some actions that require strictly symbolic information – for example, to stop a car in response to a red traffic light – which fall outside the realm of a perception-action cycle. Although such cognitive responses are not necessarily a goal of artificial life, they must necessarily be included within the embodied paradigm if it is to encompass the cognisant individual, the self-aware individual, or, potentially, the conscious individual. This paper will address the question, ‘can an animat appreciate art?’ Although this may seem very different to the example of a prosaic response to a traffic light, it will be argued that a common framework for establishing the meaning of an object is needed. It will also be argued that clarification to previous philosophical models of artistic engagement is required: in particular that the process of understanding is not a dialogue between an autopoietic artwork and animat, but that there is either a unity of object (artworkanimat) which becomes self-maintaining, or a more classical Gibsonian interpretation as a fixed set of affordances offered by an object to the subject, both of which lead to the conclusion that the process of understanding becomes a resonance in the unity or animat.

Key words: Autopoiesis, hermeneutics, embodiment, phenomenology, direct perception, natural vision