The Relationship Between Social and Cognitive Development

Willis F. Overton

• Publisher: Psychology Press
• Year: 2013
• Topic: Psychology
• eBook ISBN:
  9781134921416
• Subtopic: Developmental Psychology
• Index: Psychology

Published in the year 1983, The Relationship Between Social and Cognitive Development is a valuable contribution to the field of Developmental Psychology.

1 Empathy, Guilt, and Social Cognition

Martin L. Hoffman

University of Michigan

For several years I have been putting together a theoretical model of empathy, especially the empathic response to another person's pain or distress, and guilt, which are two of the major social affects. There is empirical as well as theoretical justification for thinking of empathy and guilt as prosocial affects, as the research—whether anecdotal, correlational, or experimental—indicates that arousing empathic distress or guilt predisposes a person to help others. What seems appropriate in view of the theme of this volume, is first to summarize the latest version of my theory of empathic development, making more explicit than in the past the rationale and connection between affective and cognitive processes. Secondly, I discuss a type of guilt that relates to empathy and highlights the interaction between affect and cognition and present a speculative account of how it develops. Finally, I discuss the implications of empathic affect for social cognition. The motivational aspect of empathy and guilt—the fact that they both include a disposition to act altruistically or morally to others—which has been heavily researched and is reviewed elsewhere (Hoffman, 1982b), is mentioned here only in passing.

EMPATHY

Empathy has been defined over the years in many ways, most of which reduce to two basic categories: empathy as cognitive awareness of another person's inner states (thoughts, feelings, perceptions, intentions); and empathy as a vicarious affective response to others, that is, an affective response that is more appropriate to someone else's situation than one's own. My primary interest is in the second definition, empathy as vicarious affect, although, as will be made clear, cognition cannot be ignored.

There is general agreement among the few writers on the topic that affective empathy is largely involuntary. Consider the following response made by one of Bandura and Rosenthal's (1966) subjects, who was given a strong dose of epinephrine before observing someone being administered a series of electric shocks:

After the first three or four shocks, I thought about the amount of pain for the other guy. Then I began to think, to minimize my own discomfort. 1 recall looking at my watch, looking out the window, and checking things about the room. I recall that the victim received a shock when I was thinking about the seminar, and that I didn't seem to notice the discomfort as much in this instance.

This quotation illustrates the involuntary dimension of empathic arousal. It also illustrates the empathic discomfort produced in the observer and the use of defensive strategies that serve to reduce the discomfort. In general, people find it difficult to avoid empathizing, especially with someone in pain or discomfort, unless they engage in perceptual or cognitive strategies like turning away from the victim or thinking about other things. If one pays attention to the victim, one will ordinarily respond empathically. The reason for this is that very simple, primitive psychological mechanisms appear to underlie empathy. Let me now describe these mechanisms.

Vicarious Affect Arousal

There appear to be at least five distinct modes of empathic arousal, which vary in degree of perceptual and cognitive involvement, type of eliciting stimulus (e.g., facial, situational, symbolic), and amount and kind of past experience required. They are here presented roughly in order of their appearance developmentally.

1. Primary Circular Reaction and the Reactive Newborn Cry. It has long been known that infants cry to the sound of someone else's cry. The first controlled study of this reactive cry was done by Simmer (1971), who found it in 2- and 3-day-olds. He also established that the cry was not simply a response to a noxious physical stimulus, as the infants did not cry as much to equally loud and intense nonhuman sounds. There thus appears to be something especially unpleasant about the sound of the human cry. Simner's findings have been replicated in 1-day-olds by Sagi and Hoffman (1976), who report in addition that the subject's cry is not a simple imitative vocal response lacking an affective component. Rather, it is vigorous, intense, and indistinguishable from the spontaneous cry of an infant who is in actual distress. No one yet knows the reason for this reactive cry, although I can see three possible explanations: (a) it is innate; (b) it is a primary circular reaction: Because the infant cannot tell the difference between the sound of its own cry and the sound of someone else's cry, the sound of someone else's cry produces a cry response in the infant. The infant then cries to the sound of its own cry, and so on; (c) the sound of the infant's cry is associated with its own past distress—perhaps at birth. Consequently, the sound of the infant's cry serves as a conditioned stimulus for its own cry response. Because of the confusion between its own and someone else's cry, the sound of someone else's cry may also serve as a conditioned stimulus for the infant's cry response.

Regardless of which explanation is correct, the fact remains that infants respond to a cue of distress in others by experiencing distress themselves. This reactive cry therefore must be considered as a possible early, rudimentary precursor of empathy, though not a full empathic response, because it lacks any awareness of what is happening. It is also possible that the reactive cry may actually contribute to empathic distress later on because the co-occurence of distress cues in others and actual distress in self may lead to an expectation of distress in self when one perceives distress cues in others. This leads to the next mode.

2. Classical Conditioning. The second mode requires more perceptual discrimination capability and therefore appears a bit later developmentally than the reactive newborn cry. It is a type of direct classical conditioning of empathy that results from the experience of observing the distress of another person at the same time that one is having a direct experience of distress. The result is that distress cues from others become conditioned stimuli that evoke feelings of distress in the self. Aronfreed and Paskal (1966) created the necessary co-occurrence of distress in the child and distress cues from someone else by subjecting the child to an aversive sound at the same time that the experimenter (who presumably heard an even more aversive sound through earphones) showed a pained expression. This may seem like a contrived way to achieve the needed co-occurrence between distress in self and distress cues from another but it has a great deal in common with something that may often occur in real life, namely the bodily transfer of the caretaker's affective state to the infant through physical handling (Hoffman, 1978). For example, if the mother feels anxious or tense while holding the infant, the mother's body may stiffen and her distress is thus transferred to the infant. After that, the facial and verbal expressions accompanying the mother's distress may become conditioned stimuli that can evoke distress in the infant even when there is no physical contact between them. Furthermore, through stimulus generalization, when other people show similar facial and verbal expressions of distress, this may evoke a feeling of distress in the infant. In this way, cues of distress in others become conditioned stimuli for distress in the child. This mechanism may explain the behaviors fitting Sullivan's (1940) definition of empathy as a form of “nonverbal contagion and communion” between mother and infant, in which the infant is viewed as automatically empathizing with the mother, feeling euphoric when she does and anxious when she is.

3. Direct Association. A more indirect type of empathic conditioning was described some time ago by Humphrey (1922). It holds that when we observe someone experiencing an emotion, his or her facial expression, voice, posture, or any other cue in the situation that reminds us of past situations in which we experienced that emotion may evoke a similar emotion in us. The usual example cited is the boy who sees another child fall down or cut himself and cry. The sight of the child falling, the sight of the blood, the sound of the cry, or any other distress cue from the victim or the situation that reminds the boy of his own past experiences of pain may evoke the unpleasant affect associated with that experience. This mode of empathic distress arousal is not limited to distress originating in physically communicated tensions or confined to early infancy. Furthermore, it does not require the co-occurence of distress in self and distress cues from others. The only requirement is that the observer have past experiences of pain or discomfort. The feelings of distress that accompany these experiences are then evoked by distress cues from the victim that call up any of those earlier experiences. It is thus a far more general associative mechanism, one that may provide the basis for a multiplicity of distress experiences with which children, and adults as well, can empathize.

4. Mimicry. A fourth mechanism of empathic arousal was described seventy years ago by Lipps (1906). For Lipps, empathy is an innate, isomorphic response to another person's expression of emotion. There are two steps: (1) the observer automatically imitates the other with slight movements in facial expression and posture (“motor mimicry”); (2) this then creates inner kinesthetic cues in the observer that contribute (through afferent feedback) to the observer's understanding and feeling the same emotion. This conception of empathy has been neglected in the literature probably because it seemed too much like an instinctive explanation. There is some recent research (reviewed by Hoffman, 1978), however, suggesting its plausibility.

5. Symbolic Association. The fifth mode, like the third, is based on the association between the victim's distress cues and the observer's past distress. In this case, though, the victim's distress cues evoke empathic distress not because of their physical or expressive properties but because they symbolically indicate the victim's feelings. For example, one can respond empathically to someone as a result of seeing a picture of that person in an emotional situation, reading a letter from him or her, or hearing someone else describe what happened to him or her. The empathy-eliciting cue can be an emotional label (“I'm scared”) or a description of the event (“My father was taken to the hospital”). This is obviously a relatively advanced mode of arousal as it requires the ability to interpret symbols and process information semantically. The semantic processing may put distance between the observer and the other. It may also reduce the involuntary component, though not totally because the symbols serve primarily as the medium by which the other's feeling is communicated; if one understands how the other feels, one can usually be expected to respond empathically.

6. Role Taking. In the sixth mode, which is clearly the most advanced developmentally, the person imagines how it would feel if he or she were in the other's place. The pertinent research has been done mainly by Stotland and his associates. In one study (Stotland, 1969), a group of subjects were instructed to imagine how they would feel and what sensations they would have in their hands if exposed to the same painful heat treatment that was being applied to another person. These subjects gave more evidence of empathic distress both physiologically and verbally, than (a) subjects instructed to attend closely to the other person's physical movements and (b) subjects instructed to imagine how the other person felt when he or she was undergoing the treatment. The first finding is not surprising. It suggests simply that imagining oneself in the other's place is more empathy arousing than observing the other's movements. The second finding, however, may tell us something about what is actually imagined. Specifically, it suggests that empathic affect is more likely to be generated when the direct focus of attention is not on the model's feelings but on the model's situation and on how one would feel in that situation, that is, when one imagines how one would feel if the stimuli impinging on the model were impinging on oneself. From a purely cognitive standpoint, the process may resemble that in which we imagine how we would feel in a hypothetical situation having nothing to do with another person. The difference of course is that when we know that we are responding to what is happening to someone else rather than to what might happen to us, we have the subjective experience of feeling what the other person is feeling—the feeling that we are empathizing.

These six modes of empathic arousal do not form a stage sequence in the sense of each one encompassing and replacing the preceding. The first mode typically drops out after infancy, owing to controls against crying, although even adults feel sad when they hear a cry and some adults even feel like crying themselves, though they usually control it. The sixth mode, being deliberate, may be relatively infrequent—used at times, for example, by parents and therapists who believe it is important to feel something of what the child or patient is feeling. The intermediate four modes, however, enter in at different points in development and may continue to operate through the life span.

Which of these arousal modes operates in a given situation presumably depends on the cues that are salient. If the expressive cues from the victim are salient, then mimicry may be the predominant mode. Conditioning and association are apt to predominate when situational cues are salient. Cues based on pictorial or verbal communication will of course require symbolic association. And, in any of these cases, the possibility exists for additional arousal if the observer gives thought to how he or she would feel in the other's situation. In other words, an arousal mode exists for whatever type of cue about the other's feelings may be present, and multiple cues may increase the level of arousal. This is important because it indicates that empathy may be an overdetermined response in humans. Empathy may also be a self-reinforcing response. For example, as already mentioned in connection with the newborn's reactive cry, every time we empathize with someone in distress the co-occurrence of our own distress and distress cues from another person may increase the strength of the connection between the cues of another's distress and our own empathic response, thus increasing the likelihood in the future that distress in others will be accompanied by empathic distress in ourselves. Another aspect of these arousal modes worth noting is that most of them require rather shallow levels of cognitive processing (e.g., sensory registration, simple pattern matching, conditioning) and are largely involuntary. With several such arousal modes in the human repertoire, the evidence that empathy may be a universal and largely involuntary response—if one attends to the relevant cues one responds empathically—that had survival value in human evolution should perhaps not be surprising (Hoffman, 1981a).

Development of a Cognitive Sense of Others

Before discussing the cognitive component of empathy I should note that cognition did enter into the discussion of arousal in several ways. First, imagining oneself in the other's place, as in the most advanced mode, illustrates the role of cognition in actually generating empathic affect. Second, even in the earlier arousal modes, perceptual discrimination and pattern recognition must often mediate between the facial and situational cues of the model and the affect aroused in the observer. The role of cognitive mediation is especially apparent when the observer responds to the symbolic meaning of stimuli rather than to their physical attributes, as in the fifth mode. What I mean by the cognitive component, however, is something more fundamental and unique to empathy.

Though empathy may usually be aroused by the predominantly simple, involuntary mechanisms just described, the subjective experience of empathy is rather complex. The literature on empathy often stops with the idea that the observer feels vicariously what the model feels through direct experience. This is the essential feature of empathy, to be sure, but mature adults who observe another person in pain, for example, know that their own distress feeling is due to a stimulus event impinging on someone else. They also have an idea of what the other person is feeling—an idea based on how they would feel in the other's situation, their general knowledge about how people feel in such a situation, or their specific knowledge about that other person. Young children who lack the distinction between self and other may have vicarious affect aroused without these cognitions. In other words, how people experience empathy depends on the level at which they cognize others. This suggests that empathy development must correspond at least partly to the development of a cognitive sense of others. The cognitive sense of others undergoes dramatic changes developmentally and thus provides a conceptual basis for a developmental scheme for empathy.

Although extensive work has been done on role taking, there is as yet no formal literature on a broader conception of a cognitive sense of the other. I have worked out such a conception, based on several different bodies of research, which will now be summarized.

Person Permanence. Person permanence refers to the awareness of another's existence as a separate physical entity. The young infant apparently lacks this awareness; objects, events, and people are not experienced as distinct from the self. Not until about 6 months, according to Piaget (1954), does the infant organize the fleeting images making up his or her world into discrete objects and experience them as separate from his or her own biologically determined sensations. The main empirical evidence comes from studies of object displacement. If a desired object is hidden behind a screen before the infant'...

Table of contents

• Cover
• Half Title
• Full Title
• Copyright
• Contents
• Preface
• 1. Empathy, Guilt, and Social Cognition
• 2. Domains and Categories in Social—Cognitive Development
• 3. Cognition of Physical and Social Events
• 4. The Nature of Social—Cognitive Change in the Developing Child
• 5. Who Killed the Epistemic Subject?
• 6. Children's Understanding of Multiple Emotions: A Cognitive—Developmental Approach
• 7. Notes on the Cognitive—Developmental Approach to the Study of Social Cognition in Children
• 8. Piaget and the Self Constituted through Relations
• 9. Piaget, Play, and Cognition, Revisited
• 10. Sutton-Smith, Play, and Maybe Piaget
• Author Index
• Subject Index