This fifth Emory Symposium in Cognition is both like and unlike its predecessors. It resembles them in its recognizably ecological emphasis, as well as in its attempt to integrate the study of adult cognition with that of development. Unlike them, however, its scope is not defined by a traditional field like “memory” or “categorization.” Instead, it is delineated by a more fundamental distinction, that between the self and everything else. William James said it best: “Each of us dichotomizes the Kosmos at a different place” (1890: 290). As perceivers and thinkers, all human beings learn things about both sides of that dichotomy. We perceive, remember, and think about the world around us, but also about ourselves. This book, the first of three projected volumes, deals with perceiving. It focuses on two forms of self perception, the ecological and the interpersonal A companion volume on the remembering self follows soon; a third, still in preparation, will focus on self-concepts.

The self may be a new topic for cognitive psychologists, but not for everyone. Philosophers and psychologists of various persuasions have written about it for many years with considerable insight.

Abstract: This target article draws together two groups of experimental studies on the control of human movement through peripheral feedback and centrally generated signals of motor commands. First, during natural movement, feedback from muscle, joint, and cutaneous afferents changes; in human subjects these changes have reflex and kinesthetic consequences. Recent psychophysical and microneurographic evidence suggests that joint and even cutaneous afferents may have a proprioceptive role. Second, the role of centrally generated motor commands in the control of normal movements and movements following acute and chronic deafferentation is reviewed. There is increasing evidence that subjects can perceive their motor commands under various conditions, but that this is inadequate for normal movement; deficits in motor performance arise when the reliance on proprioceptive feedback is abolished either experimentally or because of pathology. During natural movement, the CNS appears to have access to functionally useful input from a range of peripheral receptors as well as from internally generated command signals. The unanswered questions that remain suggest a number of avenues for further research.

Introduction

All mammals exhibit some form of awareness: Attention, proprioception, the capacity for pain, as well as a sense of permanence, agency, and continuity constitute “the machinery of the self” (Lewis, SAAH2), which underlies simple awareness. At least some aspects of the machinery of the self are probably universal among animals because they are necessary for goal-directed behavior by any kind of entity (Fehling, personal communication). Selfawareness, in contrast, goes beyond the machinery of the self to the idea of me (Lewis, SAAH2). As the phrase “the idea of me” implies, self-awareness involves cognitive and affective components that exceed those necessary for simple awareness. In this paper we are concerned first with self-awareness and how it develops in human children. Second, we are concerned with the implications that developmental models of self-awareness may have for understanding self-awareness in our closest relatives, great apes, lesser apes, and Old World monkeys.

Drawing on William James's (1892/1961) famous distinction between the subjective and objective self, we understand subjective self-awareness in human adults to be an individual's awareness of his own bodily, emotional, mental, and social characteristics, his own goals, plans, and intentions, and of the strategies he might employ in the service of those goals.

Introduction

In this volume and in related publications, various authors have presented a rich and heterogeneous array of hunches and hypotheses about the possible mechanisms of self-awareness, and their ontogeny and evolution. Some of these are full-blown theories, others are presented merely as intuitions. In this last chapter we take on the task of identifying, classifying, and comparing these diverse and often contradictory models. For clarity, we organize our review and evaluation, following Tinbergen (1963) and others (Hinde, 1982; Yoerg & Kamil, 1990), according to four approaches to the study of evolved behavioral capacities: their proximate or immediate causation (physiological, information processing, or psychological mechanisms or organizing principles), ontogeny, specific evolutionary history in a particular lineage, and adaptive functions in various species. A full evolutionary explanation should embrace all four approaches.

These four approaches focus on complexly interrelated facets of behavior: The proximate mechanisms and ontogeny of behaviors in each lineage have been favored by a specific history of selective and random forces and developmental constraints operating on a set of preexisting mechanisms and developmental patterns. Specific mechanisms and their development have been favored because they met certain selection pressures present in other members of that lineage at the time these mechanisms evolved. Similar functional systems based on different mechanisms and developmental patterns have often evolved independently in distantly related species with differing histories and distinct preexisting complexes because they met similar selection pressures.

Abstract: It is increasingly clear that spinal reflex systems cannot be described in terms of simple and constant reflex actions. The extensive convergence of segmental and descending systems onto spinal interneurons suggests that spinal interneurons are not relay systems but rather form a crucial component in determining which muscles are activated during voluntary and reflex movements. The notion that descending systems simply modulate the gain of spinal interneuronal pathways has been tempered by the observation that spinal interneurons gate and distribute descending control to specific motoneurons. Spinal reflex systems are complex but current approaches will continue to provide insight into motor systems. During movement, several neural mechanisms act to reduce the functional complexity of motor systems by inhibiting some of the parallel reflex pathways available to segmental afferents and descending systems. The flexion reflex system is discussed as an example of the flexibility of spinal interneuron systems and as a useful conceptual construct. Examples are provided of the kinds of experiments that can be developed using current approaches to spinal interneuronal systems.

Introduction

The theory of interpersonal communication and its motives

In this chapter I shall analyze the condition of a human self around the time of birth, as seen when a baby is in communication with an other. Then I shall trace some important changes that have been recorded in the life of the communicating self through infancy.

Communication with persons is possible from birth, and we should not be surprised at this. It is in the nature of human consciousness to experience being experienced: to be an actor who can act in relation to other conscious sources of agency, and to be a source of emotions while accepting emotional qualities of vitality and feeling from other persons by instantaneous empathy. This interpersonal self, this person who breaks the private integrity of the ecological self, splitting its egocenter and reconstituting it as a part of a communication dipole or multipole, is fundamental to the human condition.

Nevertheless, our psychological tradition, observing the limited cognitive powers of the newborn and giving great value to the cultivated intelligence of an educated person, has assumed that the mind of an infant is incoherent, with undefined perceptions and incapable of contributing to communication, except to solicit help reflexively for biological functions.

The title of this chapter may seem like an assemblage of disparate terms, but I hope to show that they are all interrelated. My method of analysis is phenomenological – that is, it seeks to unravel the structure and values of a phenomenon, in this case, self-knowledge in dance, through reductive analysis. In order to avoid being sidetracked with an explanation of phenomenology, I will simply say that phenomenology is a descriptive method; it is also philosophical. Once embarked upon a phenomenological description, one is obligated to pursue the essence of the phenomenon, the thing in itself. Phenomenology aims to describe and define through a distillation of consciousness. In this sense, it is reductive. In focusing on self-knowledge in terms of dance, agency and freedom arise as important experiential values. They are intrinsic values in dance, aspects of the dancer's knowledge of, or perception of, self. Agency and freedom are interrelated aspects of dancing moments (those times when our intentions are realized in dance), and dancing moments are the basis for self-knowledge in dance. I will utilize the psychological concept of the “ecological self,” from Ulric Neisser, the self directly perceived as it moves through, acts upon, and experiences its immediate environment.

Several years ago, I used the phrase “the anomalous gorillas” while reviewing the nonhuman primate self-recognition literature (Povinelli, 1987, p. 496). At that time, there had been a number of careful, unsuccessful attempts to produce a convincing demonstration of self-recognition in this species (Ledbetter & Basen 1982; Lethmate, cited in Gallup, 1982; Suarez & Gallup, 1981). However, there was one gorilla (Koko) who, based upon uncontrolled (at the time) observations, appeared to show clear signs of self-recognition. The descriptions of Koko's behaviors offered by Patterson (1984) were compelling and comparable to that obtained by Gallup and his colleagues, as well as by other independent researchers working under more controlled conditions with chimpanzees (Calhoun & Thompson, 1988; Gallup, 1970; Gallup, McClure, Hill, & Bundy, 1971; Lethmate & Dücker, 1973; Suarez & Gallup, 1981). Despite this rather remarkable behavior on the part of one member of the species, gorillas as a whole appeared to fall into a different category than chimpanzees, orangutans, and humans with regard to self-recognition. This seemed difficult to reconcile with the evolutionary relatedness of the species within the great ape–human clade; hence my invocation of the phrase “the anomalous gorillas” (Povinelli, 1987).

Several potential explanations for why most gorillas did not appear like chimpanzees on mirror self-recognition tasks suggested themselves, ranging from the hypothesis that they are differentially sensitive to the effects of being reared in captivity, to the possibility that they are simply not motivated to solve the learning problem imposed by the introduction of mirrors (Povinelli, 1987).

Abstract: The premise explored in this target article is that the function of the cerebellum is best understood in terms of the operation it performs across its structurally homogeneous subdivisions. The functional heterogeneity sometimes ascribed to these different regions reflects the multiplicity of functions subserved by the central targets receiving the outputs of different cerebellar regions. Recent studies from our own laboratory and others suggest that the functional unit of the cerebellum is the sagittal zone. It is hypothesized that the climbing fiber system produces a short-lasting modification in the gain of Purkinje cell responses to its other principal afferent input, the mossy fiber-granule cell-parallel fiber system. Because the climbing fiber inputs to sagittally aligned Purkinje cells can be activated under functionally specific conditions, these afferents could select populations of Purkinje neurons that would be most highly modulated by mossy fiber inputs responding to the same conditions. These operations may be critical for the on-line integration of inputs characterizing external target space with features of the intended movement, proprioceptive and kinesthetic cues, and the body image.

Introduction

An examination of Western philosophical and psychological perspectives on the self reveals that the self can be viewed as a soul, an underlying substrate, an activity, an explanatory hypothesis, a cognitive structure, a verbal activity, an experience, a process, or a normative attainment (Levin, 1992, pp. 204–205). These various perspectives on “the” self present a problem for a unitary view of the self, but in fact the variety is even greater than this. According to some, multiple selves are simultaneously present in dissociative phenomena such as hypnosis or multiple personality disorder (E. Hilgard, 1986/1991), as well as in the play and exploration of young children (Wolf, 1990). According to others, the notion of multiple selves should not be taken literally, either because it stops us from recognizing that a single “person” is in control (Elster, 1986) or because it reifies what are actually only “parts” of a mythical unit (Orne & Bauer-Manley, 1991). Although many agree that there is a distinct number of types or aspects of self, they disagree as to the exact number: Self is viewed as single (Gallup, 1985), double (Lewis & Brooks-Gunn, 1979), triple (Freud, 1923/1962; Lichtenberg, 1975; Schafer, 1968), quadruple (Stern, 1985), quintuple (Neisser, 1988), septuple (Stern, 1985, p. 7), or octuple (Allport, 1961).

To know oneself is to know oneself as a person among others. In order to acquire a developed concept of self, therefore, children need to appreciate the nature of persons and to recognize the existence of other selves with whom they have much in common but from whom they are differentiated. A principal concern of this chapter is to consider how very young children come to apprehend their commonality with other people and grasp the ways in which different selves occupy distinct psychological orientations toward an objectively existing world. I shall emphasize how interpersonal communication of feeling is especially important for individuals' experience of connectedness with others, and suggest that vision plays a special but not indispensable role in enabling young children to discover the fact that different people have different attitudes and points of view vis-à-vis a shared environment that includes the children themselves.

My second concern is with young children's developing capacity to symbolize. To symbolize is to treat one thing in such a way that it stands for another. This statement is a crude oversimplification, but let that be.

Both mirror self-recognition (MSR) and “true” or “representational” imitation are among a host of different abilities that emerge in human infants between 18 and 24 months of age, when mental representation develops (Piaget, 1952). Mental representation is the highest cognitive achievement of the sensorimotor period. We believe that this is the basis for the expression of both self-recognition and imitation; by considering comparative and developmental evidence we shall discuss some of the cognitive conditions for each. Several researchers have suggested that the ability to imitate is a necessary condition for the development of self-recognition in human infants (e.g., Baldwin, 1903; Kaye, 1982; Lewis & Brooks-Gunn, 1979; see also Gopnik & Meltzoff, SAAH10; Hart & Fegley, SAAH9). However, contrary to this position, one hypothesis of this chapter is that imitation is not necessary for the development of self-recognition.

We believe that the developmental and comparative approaches must be used together in order to specify cognitive conditions for self-recognition and imitation (Mitchell, 1987; Parker, 1990, 1991). It is particularly difficult to determine the cognitive prerequisites by studying children alone, because many complex behaviors appear concurrently within a very narrow time frame. Developmental study of closely related primate species may be required to isolate the factors that are important in the development of cognitively complex behaviors such as self-recognition and imitation (Antinucci, 1989; Parker, 1991). Therefore, we shall utilize developmental data collected from our closest genetic relatives, chimpanzees, to compare with developmental data from human primates.

Like all animals, we exist by virtue of coupling our bodies to the environment through action. Inadequate coupling, whether it be on the perceptual side, as when driving in fog, or on the movement side, as caused by muscular dysfunction, can be a prescription for disaster. Perception is necessary for controlling movement just as movement is necessary for obtaining perceptual information. Perception and movement compose a cycle that is action.

Action in the environment is the root of the ecological self (Neisser, 1988). As Neisser has so neatly put it in this volume: We are what we do. Therefore, to understand the self, we need to understand how actions can be geared to the environment. The first point to note is that control has to be prospective. If we see a person or animal going around bumping into things, we think something clearly must be wrong – so much do we take prospective control for granted. For to avoid colliding with objects, the consequences of continuing one's present course of action (such as heading in a particular direction, braking with a particular force) must be perceived and evasive action taken in time.

The investigation of animal cognition is currently an active area of research in comparative psychology. Such cognitive processes as representation of lists (Swartz, Chen, & Terrace, 1991), complex strategies of spatial memory (Cook, Brown, & Riley, 1985), use of numbers (Boysen & Berntson 1989), and acquisition of concepts (D'Amato & Van Sant, 1988; Herrnstein & Loveland, 1964; Oden, Thompson, & Premack, 1990) have been addressed in primate and nonprimate species. One of the most exciting topics relevant to both comparative and developmental psychology has been that of mirror selfrecognition (MSR), with its implications for comparative cognition (see Gallup, 1982, 1991; Gallup & Suarez, 1986; Mitchell, 1993; Parker, 1991; Swartz, 1990).

This phenomenon, first demonstrated in chimpanzees by Gallup (1970), involves two empirical observations that have been interpreted as evidence of self-recognition. When first exposed to a mirror, chimpanzees showed socially appropriate behaviors directed toward the mirror, as though the reflection were another conspecific. As mirror exposure continued, the chimpanzees' behavior changed. Social behaviors waned and self-directed or selfreferred behaviors appeared. Self-directed behaviors involve using the mirror to guide movements to areas of the body not visible without the aid of the mirror; self-referred behaviors involve manipulation of the environment while watching the action in the mirror. The appearance of self-directed behaviors was interpreted as indicative of self-recognition on the part of the chimpanzee (Gallup, 1970).

Self-awareness, and the object of its reflection, the self-concept, are central features of the conscious experience of adult humans in Euro-American cultures. The phenomenal reality of these facets of experience have fascinated scholars interested in human nature for centuries. Philosophers, psychologists, sociologists, anthropologists, and literary critics have offered (often contradictory) perspectives on the essence of the sense of self. Perhaps the single most perplexing issue in the vast literature concerns the genesis of selfawareness and the self-concept: How can the zygote – which by consensus is without self-awareness or a self-concept – develop into an adult human within whom these qualities are apparent?

Many of developmental psychology's greatest theorists (e.g., Baldwin, 1902; Guillaume, 1926/1971; Luria, 1976; Mead, 1934; Piaget, 1932/1965) have attempted to explicate the nature of self-awareness and the self-concept, as well as the processes through which they are acquired. There is agreement among these early theorists that social interaction of one type or another (imitation, role taking, and language use have all been proposed) is central to the development of the sense of self. (For a partial review of these theories, see Hart, Kohlberg, & Wertsch, 1978.) The postulate of the social origins of self-awareness and the self-concept continue to be evident in more recent accounts as well (e.g., Mahler, Pine, & Bergman, 1975; Stern, 1985).

Introduction

Laboratory studies of mirror self-recognition (MSR) using Gallup's controlled method of face marking under anesthesia (Gallup, 1970) have yielded positive results for chimpanzees and orangutans, but negative results for gorillas (Ledbetter & Basen, 1982; Suarez & Gallup, 1980). In contrast, both observational and informally controlled studies of the language-trained gorilla Koko have yielded positive results (Patterson & Cohn, SAAH17). In this chapter, I report observations of mirror-stimulated behaviors in a group of gorillas at the San Francisco Zoo and in a group of chimpanzees at the Oakland Zoo, which support the hypothesis that gorillas in general are capable of MSR.

Observations on gorillas

At the time of my first observations (in 1978) the gorilla group at the San Francisco Zoo was composed of an adult silverback male, Bwana; his 5-yearold son, Sunshine; his 3-year-old son, Mkumbwa; Sunshine's mother, Mrs.; Mkumbwa's mother, Jacqueline; and an adult nulliparous female, Pogo. During these observations, the animals were moving freely in and out of their night cages where they could see a 15 × 60–inch mirror I had set up outside the cage. During the hour that they had access to the mirror, I took ad-lib notes on any animal that was looking in the mirror.

On this occasion, the female Pogo inadvertently marked her own face with fingerpaint when she put her face down on the cardboard she had been marking. She subsequently looked in the mirror and wiped the paint off with her hand.