8. 10 Restatement of Themes in this Chapter

Widespread metaphors that liken media to channels for conveying messages do not lend themselves well to explanations of how human beings interact with mediated representations. Nor can they entirely explain how mind and media interact with each other to generate complex and dynamic cognitive phenomena that neither mind nor media can alone support.

Thinking of media as channels for sending and receiving symbols has often led by extension to the conclusion that perception is a process of reception and that cognition is the processing of symbols and language-based propositions. However, many interactions of mind and media are not easily explained in terms of channels and symbols, because humans, like most organisms, understand their environments through exploratory action and active perception. Successful design of products and services for emergent media technologies will depend in part on the extent to which such products and services honor modalities of integrated action-perception grounded in the necessities of survival and reproduction and tuned to opportunities for action in ecological niches.

Ecological psychology assumes that thermodynamic laws govern the structure and function of living communities and that successful strategies for living must be in accord with these laws. An organism's thermodynamic efficiency is partly a function of the means by which the organism obtains information about its environment and the strategies by which it uses this information to influence or control energy and matter. Living communities generate "organized complexity" by leveraging relatively small amounts of information to influence or exploit larger flows of energy and matter. Information storage and processing in turn require energy and matter, and this requirement can be thought of metaphorically as a cost to organisms, or as an investment.

The most widespread and common systems for storing and processing information on this planet are based on DNA. Yet DNA-based information systems are fundamentally limited in capacity and rate of evolution. While nervous systems offer the advantage of greatly increased flexibility for storing and processing information about the environment, such systems also impose biological costs. Off-loading information storage and processing to systems that, are external to the organism can reduce this cost.

From the standpoint of cognitive theory, the environment can be considered a generator of information about itself, and perception the means of obtaining this information. In general, natural selection appears to favor investment of resources by a species in systems for detecting, selecting, and perceiving information in the environment when the cost of obtaining such information is less than the cost of generating or obtaining equivalent information from sources internal to the organism. Investing resources in improved perception is a means of obtaining information about the environment while minimizing investment in biological mechanisms for internal information processing.

The most important information an organism can derive from perception is information about invariant or unchanging aspects of the environment, for example, gravitational fields. Invariants anchor perception and action by functioning as "knowns" in decision-making or problem-solving processes. Action-perception related to invariants is often highly automatic, and in human beings is typically processed unconsciously.

Natural selection tunes perceptual processes to affordances or opportunities for action associated with an organism's ecological niche or occupation. Since selection similarly tunes effectivities or capabilities for action to opportunities, perception in most organisms is strongly associated with action. Indeed, a fundamental tenet of ecological psychology is that divorcing the study of perception from the study of action leads to a distorted understanding of attentional and intentional processes in organisms. Although cybernetic models often treat the relationship between action and perception as cyclic, a more appropriate alternative in many cases is to treat this relationship as covariant or codeterminant.

Human evolution has been accompanied by an increasing reliance on external information storage and processing. Social routines and gestures probably allowed early human beings to "off-load" much of their individual information processing to external venues associated with group activity--sharing the work of cognition with artifacts as well as with other human beings. However, it seems unlikely that human beings used language in the form of high-speed articulate speech to externalize information until the last few hundred thousand years. External storage became even more stable and reliable with the advent of markings, glyphs, and alphabets.

Modem minds can be arguably characterized as much by their dependence on tools and artifacts as by any purely internal mental process. Indeed, it appears that thinking and reasoning in today's artifact- and media-rich societies are best viewed as emergent functions of distributed cognitive systems in which the work of information storage and processing is shared between realia, media, and Mental-Internal Representations of Situations (MIROS).

It seems reasonable, therefore, to assume that much everyday cognition and many important modalities of thought are governed not by purely internal mental models of the way the world works but rather by open models that require a constant flux of data about how action is related to perception. Much conventional thinking about media technologies has been strongly influenced by traditions of empirical research that have divorced the study of perception from the study of action. These traditions treat perception as the processing of symbols and cognition as the analysis of propositions. Yet many emerging computer environments for learning, work, and play invoke modalities of action-perception in which dispositional and enactive properties of objects take precedence over the purely symbolic meaning of such objects. Typically, these environments offer opportunities for action in which users realize their intentions by manipulating objects rather than by constructing language-based commands. Some of the most powerful and effective strategies for using interactive media, however, employ mixed modalities in which language and symbol-based communication operate in concert with object-oriented manipulation. Although older media formats such as print and cinematography do not support a high degree of object manipulation, these formats invoke human capacities for integrating perception and action by selectively substituting perception components for action components. Verbal descriptions, for example, can serve as surrogates for action by informing viewers of how and where photographic images were captured. By reducing the action component, photographs create opportunities for reflection and deliberation that may not have been available to observers at the original scene. In contrast, cinematographs substitute the dynamics of camera movement for the real or imaginary actions of observers. Such dynamism may suppress reflection and deliberation. As we acknowledge media for their potential as arenas

for action, as means for exploring the relation between acting and perceiving, researchers and designers will begin to address strategies for modulating and varying these opportunities in order to support specific educational purposes and functions. Yet only by respecting our fundamental evolutionary heritage as organisms whose cognitive capacities are grounded in ecological necessity can we hope to build media environments that allow us to live well as human beings.