Global Perspectives on the Ecology of Human-Machine Systems

John M. Flach, Peter A. Hancock, Jeff Caird, Kim J. Vicente, John M. Flach, Peter A. Hancock, Jeff Caird, Kim J. Vicente

• Publisher: CRC Press
• Year: 2018
• ISBN:
  9781351444644
• Edition: 1
• Topic: Tecnologia e ingegneria
• Subtopic: Salute e sicurezza sul lavoro

There is a growing consensus in the human factors/ergonomics community that human factors research has had little impact on significant applied problems. Some have suggested that the problem lies in the fact that much HF/E research has been based on the wrong type of psychology, an information processing view of psychology that is reductionistic and context-free. Ecological psychology offers a viable alternative, presenting a richer view of human behavior that is holistic and contextualized. The papers presented in these two volumes show the conceptual impact that ecological psychology can have on HF/E, as well as presenting a number of specific examples illustrating the ecological approach to human-machine systems. It is the first collection of papers that explicitly draws a connection between these two fields. While work in this area is only just beginning, the evidence available suggests that taking an ecological approach to human factors/ergonomics helps bridge the existing gap between basic research and applied problems.

Chapter 1

The Ecology of Human-Machine Systems: A Personal History

John M. Flach

Wright State University

Armstrong Laboratory,

Wright-Patterson Air Force Base

I first became interested in the ecological approach to psychology when, as a graduate student at Ohio State, I heard Rik Warren describe the properties of flow fields and how they might be specific to properties of locomotion such as heading, altitude, and speed. It occurred to me that these descriptions of optical structure may have far more relevance to understanding a skill, such as landing a plane, than the changes in slope or intercept of a reaction time function that were, at that time, central to the chronometric analyses of mind that dominated much of my graduate training (even though these reaction times may have been measured while the operator was simultaneously flying a simulator). As I learned more about the ecological approach it seemed obvious to me that challenges such as understanding automobile driving and flight were important to the evolution of Gibson’s theories about behavior. It was surprising to me that those interested in human factors and those interested in ecological approaches to behavior were not actively embracing each other’s theories and problems. However, I have always tended to underestimate the inertia in systems. Although the merging of ecological theories and human factors challenges has not happened as quickly as I expected, I think there is a gradually accelerating movement toward communion. This book is perhaps evidence of this movement and will hopefully be a stimulus to encourage a continuing movement toward communion.

To set the stage for this book, I would like to briefly present my personal perspective on the events that have made such a book inevitable. To begin, as I mentioned earlier, the obvious roots for a union of human factors challenges with ecological perspectives are Gibson’s early works on automobiles (Gibson & Crooks, 1938) and his work in aviation (Gibson, 1944, 1947/1958; Gibson, Olum, & Rosenblatt, 1955). However, Gibson was not alone in his insights. Langewiesche (1944), in his analysis of the information for landing and approach, anticipated many of the ideas about optical sources of information for controlling flight. Also, the need for an ecological approach, that is, an analysis whose scope is the human environment as a system, was voiced by Taylor (1957) as he described the importance of research on human-machine systems for the future of basic psychology:

“It drew attention to the fact that in many circumstances the behavior of the man was inseparably confounded with that of the mechanical portions of his environment. This meant that psychologists often could not study human behavior apart from that of the physical and inanimate world - that all along they had been studying the behavior of man-machine systems and not that of the men alone. The inseparability of the behavior of living organisms from that of the physical environment with which they are in dynamic interaction certainly argues against maintaining separate sciences and construct languages: one for the environment, the other for that which is environed.” (Taylor, 1957, pp. 257-258)

One of the earliest papers, other than Gibson’s, to analyze the information in optic flow relative to vehicular control was David Lee’s (1976) paper, “A Theory of Visual Control of Braking Based on Information About Time-to-Collision.” In this paper, Lee introduced the higher order optical variable tau. Lee wrote:

“A mathematical analysis of the changing optic array at the driver’s eye indicates that the simplest type of visual information, which would be sufficient for controlling braking and would also be likely to be easily picked up by the driver, is information about time-to-collision, rather than information about distance, speed, or acceleration/deceleration. It is shown how the driver could, in principle, use visual information about time-to-collision in registering when he is on a collision course, in judging when to start braking, and in controlling his ongoing braking.” (p. 437)

Perhaps, the first laboratory dedicated specifically to an ecological approach to problems of human-vehicular systems was not established until the late 1970s and early 1980s. During this time, the Aviation Psychology Laboratory at Ohio State was under the direction of Dean Owen. Dean assembled the components for a visual flight simulation system to evaluate the optical flow field as a source of information for flight control. During the 1980s, a number of theses and dissertations evaluated sources of information for judgments about altitude and speed. Much of this research is reviewed in a chapter written by Owen and Warren (1987) in Ergonomics and Human Factors: Recent Research (Mark, Warm, & Huston, 1987). Many of Owen’s students have gone on to continue active research careers in Aviation Human Factors. I was lucky to be on the fringes of this group during my graduate training at Ohio State in the early 1980s. This association had a major impact on my future research agenda.

The analysis of optical information specific to vehicular control continues to be one of the areas of active interchange between ecological theory and human factors problems. In 1986, Rik Warren and Alex Wertheim organized an international workshop on the perception and control of self-motion held at the Institute for Perception TNO, Soesterberg, The Netherlands. This workshop brought together a diverse group of researchers, and although many among the group would not consider themselves “Gibsonian,” the influence of Gibson’s work was clearly evident. This workshop led to the publication of a book, Perception and Control of Self-Motion (Warren & Wertheim, 1990). In Spring of 1989, Walt Johnson and Mary Kaiser hosted a workshop on Visually Guided Control of Movement at the NASA Ames Research Center, Moffett Field (Johnson & Kaiser, 1990). Many important issues with regard to the nature of information in optic flow and the importance of the coupling between perception and action were discussed at this workshop (see also Flach, 1990a).

Outside the area of vehicular control, Warren’s paper, “Environmental Design as the Design of Affordances,” presented at the Third International Conference on Event Perception and Action, Uppsala, Sweden (1985), was a landmark in the merging of human factors with ecological theory. Warren writes:

“Analyzing an affordance requires a task-specific description of an ecosystem that considers the relevant organism and environmental variables and the biomechanics of the task. The fit between organism and environment must be measured relationally, using methods of intrinsic measurement, and can be characterized in terms of optimal points at which performance is most efficient or comfortable, and critical points, at which performance breaks down.” (p. 2)

The role of intrinsic measurement as a basis for scaling the affordances in artifactual environments has become a cornerstone for building ecological theories of the workplace (see also Mark, 1987; Mark & Voegele, 1987, Warren, 1984, 1987; Warren & Whang, 1987). In 1988, innovation, which is the journal of the Industrial Designers Society of America, featured three articles that address the concept of intrinsic measurement and its importance for design (Mark & Dainoff, 1988; Rutter & Newell, 1988; Rutter & Wilcox, 1988).

Another important area of research that has seen the impact of an ecological approach is the problem of interface design for process control and decision support. Dave Woods was one of the first to bring ecological theory to bear on this problem. At a NATO Advanced Study Institute on Intelligent Decision Aids in Process Environments, Woods (1986) wrote:

“The important point for the development of effective decision support systems is the critical distinction between the available data and the meaning of the information that a person extracts from that data (e.g., S. Smith 1963). The available data are raw materials that the observer uses to answer questions (questions that can be vague or well formed, general or specific). The degree to which the data help answer those questions determines the informativeness or inferential value of the data. Thus, the meaning associated with a given datum depends on its relationship to the context or field surrounding the data including its relationship to the objects or units of description of the domain (what object and state of the object is referred to), to the set of possible actions and to perceived task goals (after Gibson, 1979, what that object state affords the observer). The process is analogous to figure-ground relations in perception and shows that information is not a thing-in-itself but is rather a relation between the data, the world the data refers to, and the observer’s expectations, intentions, and interests. As a result, informativeness is not a property of the data field alone, but is a relation between the observer and the data field.” (p. 163)

Rasmussen (1986) also references Gibsonian theory in his book, Information Processing and Human-Machine Interaction: An Approach to Cognitive Engineering. Rasmussen wrote:

“As I understand Gibson’s concept of direct perception, the “dynamic world model” is in the present context very similar to the mechanisms needed for the “attunement of the whole retino-neuro-muscular system to invariant information” (Gibson, 1966, p. 262), which leads to the situation where “the centers of the nervous system, including the brain resonate to information.” This selective resonance relies on the existence of a generic dynamic model of the environment. The implications of Gibson’s view of perception, as based on information pickup instead of sensation input, are in many ways compatible with [my own model]. To Gibson, perception is not based on processing of information contained in an array of sense data. Instead the perceiver, being attuned to invariant information in space and time in the environment, samples this invariant information directly by means of all senses. That is, arrays of sense data are not stored or remembered. They have never been received; instead the nerve system “resonates.” In my terms, the world model, activated by the needs and goals of the individual, is updated and aligned by generic patterns in the sensed information, but the idea of an organism “tuning in” on generic time-space properties is basically similar and leads to the view of humans as selective and active seekers of information at a high level of invariance in the environmental context. The subconscious dynamic world model or the attunement of the neural system leads to the situation where primitive sense data are not processed or integrated by symbolic information processes as Minsky suggests, but the generic patterns in the array of data in the environment are sampled directly by high-level questions controlling the exploratory interaction involving all senses.” (pp. 90-91)

The Fourth International Conference on Event Perception and Action at Trieste in 1987 included a symposium chaired by Sebastiano Bagnara entitled “Errors in Human-Machine Interaction.” Presenters included Mancini, Rasmussen, Reason, and Vicente. However, almost no one except for the presenters and myself attended this session. However, this insulation from the rest of the conference helped to cement a close bond between myself, Vicente, and Rasmussen which has greatly influenced my thinking about how to attack the problems of interfaces in complex systems. The ideas presented by Rasmussen and Vicente led to a paper published in the International Journal of Man-Machine Studies (1989) — “Coping with Human Errors Through System Design: Implications for Ecological Interface Design.”

It is notable that Norman in his popular book The Psychology of Everyday Things (1988) adopted Gibson’s term affordance. He wrote:

“There already exists the start of a psychology of materials and of things, the study of affordances of objects. When used in this sense, the term affordance refers to the perceived and actual properties of the thing, primarily those fundamental properties that determine just how the thing could possibly be used.” (p. 9)

A careful reading of Norman’s presentation shows that it is somewhat at odds with Gibson’s view of affordance (in a footnote, Norman clearly acknowledges the conflict between his view and Gibson’s view of affordance). Norman confuses the affordances of an object with the information that specifies the affordances. For example, he writes that “affordances provide strong clues to the operations of things” (p. 9). Clearly, Norman is moving toward an ecological approach, but there is still a strong influence of the more traditional information processing approach, in which meaning must be constructed from the clues available. This is somewhat reminiscent of Neisser’s transition from Cognitive Psychology (1967) to Cognition and Reality (1976). Just as with Neisser, this revision in thinking will likely disturb traditional cognitive psychologists, but will not be satisfactory to many already entrained in the ecological approach. This probably applies, as well, to Woods and Rasmussen’s work. However, it is encouraging to see a number of significant contributors to the development of cognitive engineering moving somewhat in the direction of an ecological approach.

Although many will strongly object, I think that this movement toward a middle (perhaps higher) ground is healthy for our science. One of the benefits of the challenge in applying psychology to problems of human-machine systems (as opposed to conducting research exclusively within narrowly defined experimental paradigms or toy laboratory worlds) is that it provides an acid test for dogmatism of any sort. More than anything these challenges teach humility and open-mindedness.

In summer 1989, a symposium was organized for the Fifth International Conference on Event Perception and Action at Miami University entitled “The Ecology of Human-Machine System.” Presenters included Stappers and Smets, Woods, Moray, Kugler, and Vicente and Rasmussen. This session was very well attended and stimulated much interesting discussion. I think that most notable were some challenges from Stappers and Smets about how to go beyond purely kinematic models of optic flow. This is particularly significant to my thesis here, as these important theoretical issues were being raised from within a department of industrial design engineering. It is important to see that applied problems are beginning to reflect back in a way that challenges our theories and advances our basic science of behavior. This is a theme that Taylor (1957) emphasized as one of the promises of engineering psychology. Several publications resulted from this symposium (Flach, 1989; 1990b; Vicente & Rasmussen, 1990).

It is important for us to realize that hypothesis testing is not the only way to validate our theories. Design of products and the success or failure of those products is another way of validating the implicit and explicit theories that guided the design. Whereas hypothesis testing emphasizes internal validity, design emphasizes external validity. Good science demands a balance between these two forms of validity. Kirlik’s chapter (this volume) discusses design as experiment. Also, I think the work of the Form Theory Group at Delft demonstrates this alternative quite well (see Smets, 1994). The key difference between their work and work of other design groups is the explicit role perceptual theory plays in their designs.

In 1990, the problems of human-machine systems were discussed at both the International Society of Ecological Psychology’s Spring meeting in Champaign-Urbana, at which Alex Kirlik’s discussion of the affordance properties of a complex helicopt...