The claim that empathy is both automatic and representational is criticized as follows: (a) five empathy-arousing processes ranging from conditioning and mimicry to prospective-taking show that empathy can be either automatic or representational, and only under certain circumstances, both; (b) although automaticity decreases, empathy increases with age and cognitive development; (c) observers' causal attributions can shift rapidly and produce more complex empathic responses than the theory allows.

Neurophysiological studies in monkeys and neuroimaging studies in humans support a model of empathy according to which there exists a shared code between perception and production of emotion. The neural circuitry critical to this mechanism is composed of frontal and parietal areas matching the observation and execution of action, and interacting heavily with the superior temporal cortex. Further, this cortical system is linked to the limbic system by means of an anterior sector of the human insular lobe.

Preston & de Waal conflate familiarity with similarity in their attempt to account for empathy. If distinguished, we may have at hand two different kinds of empathy: egocentric empathy and empathy proper.

Self-awareness is thought to be tied to processes of higher-order perspective taking including empathy. These abilities appear to be reserved for humans, great apes, and possibly, dolphins. Recent examinations reveal that both self-awareness and empathy may have origins in the right hemisphere. It is possible that, as in language, lateralization plays a key role in the development of higher-order perspective taking and self-awareness.

Two major problems exist in studying development: Similar behaviors do not need to reflect the same underlying process, different behaviors can reflect the same process; earlier behaviors do not necessarily lead to later behaviors. Empathy, rather than social contagion, is supported by different processes; contagion supported by prewired species behavior, empathy by cognitions, in particular, the cognitions about the self – a meta-representation.

The Perception-Action Model of empathy (PAM) is both sufficiently broad and sufficiently detailed to be able to describe and accommodate a wide range of phenomena – including the apparent “cold-heartedness” or lack of empathy of psychopaths. We show how the physiological, cognitive, and emotional elements of the PAM map onto known and hypothesized attributes of the psychopathic personality.

Mammalian brains contain a variety of self-centered socio-emotional systems. An understanding of how they interact with more recent cognitive structures may be essential for understanding empathy. Preston & de Waal have neglected this vast territory of proximal brain issues in their analysis.

Empathy is a developmental process whereby individuals come to understand the emotional states of others. While the exact nature of this process remains unknown, PAM's utility is that it establishes empathy along a continuum of behavior ranging from emotional contagion to cognitive forms, a very useful distinction for understanding the phylogeny and ontogeny of this important process. The model will undoubtedly fuel future research, especially from comparative domains where data are most problematic.

The comparative study of empathy should be based on the developmental taxonomy of vicarious experiences offered by the abundant literature on infants and children's cognitive, social, and emotional development. Comparative research on the topic should refer to the various kinds of empathy emerging in an orderly fashion early in human development.

Empathy represents one of the basic forms of human expression. Empathy evolved to facilitate social behavior. The perception action model, extended to empathy, is an exciting paradigm in which to undertake contemporary cognitive and comparative neuroscience. It renders the perception of events as an active affair, both when watching others, and when performing actions.

A perception-action system may underlie the mechanisms by which human speech exchange in social interaction is managed, as well as the evolutionary precursors of these mechanisms in closely related species. Some phenomena of interaction well-studied by sociologists are suggested as a point of departure for further research.

Few individuals are constitutionally incapable of showing concern for others at an early age, and malleability is possible. Individual variations will be best understood through study of the representational prerequisites of empathy in close conjunction with caregiving environments and affective underpinnings.

Only a broad theory that looks across levels of analysis can encompass the many perspectives on the phenomenon of empathy. We address the major points of our commentators by emphasizing that the basic perception-action process, while automatic, is subject to control and modulation, and is greatly affected by experience and context because of the role of representations. The model can explain why empathy seems phenomenologically more effortful than reflexive, and why there are different levels of empathy across individuals, ages, and species.

The two contrasting theoretical approaches to visual perception, the constructivist and the ecological, are briefly presented and illustrated through their analyses of space and size perception. Earlier calls for their reconciliation and unification are reviewed. Neurophysiological, neuropsychological, and psychophysical evidence for the existence of two quite distinct visual systems, the ventral and the dorsal, is presented. These two perceptual systems differ in their functions; the ventral system's central function is that of identification, while the dorsal system is mainly engaged in the visual control of motor behavior. The strong parallels between the ecological approach and the functioning of the dorsal system, and between the constructivist approach and the functioning of the ventral system are noted. It is also shown that the experimental paradigms used by the proponents of these two approaches match the functions of the respective visual systems. A dual-process approach to visual perception emerges from this analysis, with the ecological-dorsal process transpiring mainly without conscious awareness, while the constructivist-ventral process is normally conscious. Some implications of this dual-process approach to visual-perceptual phenomena are presented, with emphasis on space perception.

The author argues that the theory of a dorsal/ventral stream for visual processing can be used to reconcile the constructivist and direct perception theories. My commentary discusses neurophysiological and psychophysical studies that run counter to the view. In addition, the central issue of debate between the constructionist and direct perception approaches regarding what is visual information is discussed.

Our commentary focuses on the functional link between the ventral and dorsal systems implied by Norman, as they relate to overt movement. While issues relating to space perception and size constancy are the primary justification for this dual-process theory, the philosophical extensions of this approach are less consistent with examination of motor control and, in particular, motor learning.

The attempt to relate distinctions in perceptual theory to different physiological systems leads to numerous exceptions and inconsistencies. A more promising approach to the reconciliation of constructivist theory and direct perception is to recognize that perception does involve inference, as the constructivists insist, but that inference is a process in logic that does not require unconscious reasoning and need be no more thought-like than resonance.

The parallel drawn by Norman between the dorsal and ventral systems and direct and indirect approaches is based on two misrepresentations of the direct approach – that it is concerned only with the unconscious control of action, and that it cannot explain learning. We propose a way of understanding the visual system differences from within the direct approach.

The conceptual space< (Gärdenfors 2000) is discussed as a representation structure that connects the constructivist and the ecological vision subsystems in an operating autonomous robot based on computer vision.

Norman's identification of a ventral system embodying a constructivist theory of perception is rejected in favor of an ecological theory of perception and perceptual learning. We summarize research showing that a key motivation for the ventral-constructivist connection, percept-percept coupling, confuses perceptual and post-perceptual processes