Interpreting Perceiver Reactions to Emotional Stimuli

ABSTRACT - The expressive reactions of perceivers to emotionally expressive stimuli have been considered to be mediated emotionally and unemotionally. In order to demonstrate emotional mediation, a distinction is proposed between supplementary and complementary emotional responses. The emotional releasing function of emotionally expressive stimuli is considered from a psychobiological perspective in terms of effects on attention and attitude formation and change. Implications for ad testing are discussed.

Citation:

Jeremy D. R. Pincus (1992) ,"Interpreting Perceiver Reactions to Emotional Stimuli", in NA - Advances in Consumer Research Volume 19, eds. John F. Sherry, Jr. and Brian Sternthal, Provo, UT : Association for Consumer Research, Pages: 245-250.

Advances in Consumer Research Volume 19, 1992      Pages 245-250

INTERPRETING PERCEIVER REACTIONS TO EMOTIONAL STIMULI

Jeremy D. R. Pincus, University of Connecticut

ABSTRACT -

The expressive reactions of perceivers to emotionally expressive stimuli have been considered to be mediated emotionally and unemotionally. In order to demonstrate emotional mediation, a distinction is proposed between supplementary and complementary emotional responses. The emotional releasing function of emotionally expressive stimuli is considered from a psychobiological perspective in terms of effects on attention and attitude formation and change. Implications for ad testing are discussed.

INTRODUCTION

Some theorists have presumed the expressive reactions of perceivers to emotionally expressive stimuli to be mediated emotionally, whereas others have viewed these responses as essentially imitative. This paper attempts to resolve the affect-imitation issue by distinguishing between complementary and supplementary emotional reactions, and using this distinction as a framework for the interpretation of relevant studies. The relative contributions and validities of facial electromyography and other measures of emotion are considered. The implications of automatic affect elicitation are also considered in terms of attitude formation and change.

A major assumption of advertisers and consumer researchers has been that perceivers react emotionally to the emotional expressions of others. Emotion, here, is defined as an internal state which manifests itself through three basic output channels- physiological responses, expressive behavior, and subjective experience (Buck 1988). Accordingly, in the consumer behavior literature, the response of perceivers to emotional stimuli has been characterized as "activation" (Kroeber-Riel 1979) and an "over-time flow of feelings" (Friestad and Thorson 1986), which correspond to the concepts of physiological arousal and subjective experience of emotion, respectively. The Cohen and Areni (1991) model of affective consumer response shares this general orientation, incorporating expressive behavior in their conceptualization of the "first phase affective response". The assumption of affective perceiver reactions to emotional stimuli is clearly evident in the application of their model to the perception of a Michelin tire ad:

"The happy babies in the commercial should elicit an immediate, almost automatic, affective response. Viewers may even find themselves smiling. This 'primitive' affective response occurs without consideration of the meaning of any element of the ad. It may be more related to the facial pattern of the smiling baby's face, or the happy baby may have virtually become an 'unconditioned stimulus' for the generation of affect" (Cohen & Areni 1991, p. 232).

The idea that the emotional expressions of conspecifics cause automatic emotional reactions in observers has appeared in works as early as those of Darwin (1872) and McDougall (1908). More recently, the concept has appeared in Izard's (1977) principle of emotion contagion, and in the work of several empathy researchers (e.g. Stotland, Sherman, & Shaver 1971).

The dominant explanation of the emotion contagion process has employed the ethological concept of the innate releasing mechanism, a phylogenetically pre-programmed system for the instigation of specific motivational/emotional/ behavioral patterns. Innate releasing mechanisms are triggered by a set of stimuli known as releasers; a particular category of releaser includes the phylogenetically-based social signals of others which comprise emotional expressions. According to the ethological view, then, the emotional expressions of others act as releasers of internal emotional states in perceivers. In the words of ethologist Eibl-Eibesfeldt, "if we perceive someone smiling, cerebrochemical processes are presumably activated, these inducing a friendly mood and smiling in response" (1989, p.73). Support for this notion has come largely from studies (reviewed below) finding reliable changes in the nonverbal behaviors of perceivers upon exposure to emotional expressions (e.g. Davidson & Fox 1982; Dimberg 1988; Sackett 1966).

An alternative explanation suggests that perceiver reactions to emotionally expressive stimuli may be instances of non-emotional mimesis. Such non-emotional imitation of nonverbal behavior has itself been conceptualized as the product of an innate releasing mechanism (Masters 1979). Others have considered imitative display to serve a social communication function which may or may not be mediated by emotion (Bavelas, Black, Lemery, & Mullett 1986).

IMITATION VS. EMOTIONAL MEDIATION: RESEARCH FINDINGS

The findings of a variety of studies shed light on the imitation-emotion issue. The results of the first group of studies illustrate the ambiguity of interpretation encountered when sender and receiver both exhibit the same nonverbal display; in such cases, expressive behavior may be interpreted as either the result of emotional mediation or non-emotional imitation. These ambiguous cases will conservatively be considered supportive of the mimetic interpretation. In these studies, facial EMG is employed as a measure of emotional response, the merits and limitations of which are detailed below.

Support for Imitation

Vaughan and Lanzetta (1980) in a study of vicarious emotional conditioning, employed expressions of pain as unconditioned stimuli and collected facial EMG and skin conductance measures. Results indicated that exposure to expressions of pain, even in videotape format, can elicit congruent physiological arousal and facial patterning in the observer. The authors concluded that observers responded emotionally to the model's expressions and that the observation of emotional expressiveness is necessary for the acquisition and maintenance of vicarious conditioning. These results, however, likewise may be interpreted as an instance of imitative responding; that is, the pained expressions of the stimulus person may simply have been mimicked by observers.

McHugo, Lanzetta, Sullivan, Masters, and Englis (1985) exposed American subjects to televised emotionally expressive displays of Ronald Reagan, and compared their physiological and self-reported emotional responses. Although prior attitude significantly influenced self-reported emotion, subjects displayed similar physiological patterns of emotional expression regardless of prior attitude. Across all subjects, facial EMG responses indicated subjects smiling during displays of happiness and frowning during fear and, especially, anger displays. Skin conductance responses indicated least relaxation during anger displays, intermediate activity during fear displays, and most relaxation during happiness displays. Heart rate decelerated most during fear and anger displays. The authors concluded that emotionally expressive displays have a direct emotional impact on viewers, however, they explicitly accepted the possibility that the EMG activity may have resulted from imitation: It was suggested that the finding of "parallel" responses in sender and receiver may be interpreted in two ways, either as expressive displays mediated by internal affective state or as "motor reflexes due to the contagion of the communicator's display" (McHugo et al. 1985, p. 1527).

Dimberg (1982) demonstrated that specific patterns of facial EMG activity are spontaneously produced when subjects are presented with "positive" and "negative" facial expressions of emotion. It was found that slides of happy and angry faces elicited increased zygomatic (happy) and corrugator (negative) facial muscle activity, respectively.

Inference Based on EMG Data. Dimberg (1982) has taken these findings as evidence that facial reactions to facial expressions are emotional outputs stemming from the observer's internal state. Nonetheless, the possibility that these results are due to imitative responding cannot be ruled out. While EMG may be used to differentiate generally "positive" and "negative" reactions, further differentiation among specific "negative" emotions (i.e. anger and fear) is not easily accomplished (Ekman 1982). Therefore, when a "negative" EMG site (e.g. corrugator) is activated in response to an angry face, it is impossible to know from this information alone whether the activity is part of an anger or fear response (or sadness, disgust, or contempt, for that matter). Thus, the use of EMG to distinguish simple "positive-negative" responses to facial expressions does not permit one to unequivocally conclude that EMG responses stem from emotion.

Facial EMG remains a valuable indicator of emotional response, offering on-line, moment-by-moment tracking of changes in the activity of the facial muscles associated with emotional expression. EMG is also largely immune to subject response tendencies and biases, while being able to detect changes in muscle tone invisible to the naked eye. However, as the above analysis has shown, facial EMG results are multi-determined, both emotionally and unemotionally. A particular muscle contraction may have been caused by imitation or emotion, and if by emotion, more than one may be involved.

If the researcher is concerned only with the "positive-negative" distinction of emotion (though this conceptualization has been criticized as simplistic and confounded with evaluation; Batra and Ray 1986), facial EMG may be sufficient as a dependent measure, presuming that imitation or other confounds are not a problem. However, if the research requires a complete description of the emotional response (linked to underlying theory of emotion), it becomes necessary to employ self-reported measures and, possibly, measures of physiological response. Thus, a combination of theoretically grounded self-reported and physiological measures with facial EMG should permit reasonably valid interpretation of the emotionality of reactions.

Support for Emotional Mediation

Studies which demonstrate emotional responses to emotional expressions do so in a variety of ways. These studies have employed measures such as behavioral observation (approach/avoidance; behavior checklists), electroencephalography (EEG), and facial EMG combined with self-report.

Dimberg (1988) replicated his earlier finding of increased zygomatic and corregator activity in response to happy and angry facial stimuli, respectively, with the additional finding of corresponding self-reports of feelings of happiness in response to happy faces and feelings of fear in response to angry faces. As indicated above, this additional employment of self-report allows greater confidence in inferring that the facial responses were mediated by emotion.

Perhaps the strongest evidence of emotional responses to emotional expressions comes from Sackett (1966), who presented slides of various social and non-social stimuli to eight isolation-raised rhesus monkeys repeatedly at different points in their development beginning at 2-weeks of age and continuing up to 9 months, recording their behavior during each slide presentation. Slides of adult male monkeys displaying threat were accompanied by disturbance behavior and gaze aversion, which have been described as fear responses, while non-threatening monkey pictures were not met with these behaviors. Sackett concluded that these findings constituted evidence of an innate recognition and releasing mechanism for social communication.

Schwartz, Izard, and Ansul (1985), using a paired-comparison novelty discrimination test, presented 5-month-old infants with photos of adult women displaying anger, fear, and sadness in one condition and anger, joy, and interest in another. They found that fear and sadness were discriminated successfully; infants looked at the novel stimulus longer than at the familiar stimulus. When the anger expression served as the novel stimulus, however, discrimination could not be assessed as there was a "marked tendency for infants to look away from, or to look less at, anger" (p. 71). The authors concluded that the novelty of expressions determined visual responses less than the inherent aversiveness or social significance of the anger expression.

Davidson and Fox (1982) measured the affective responses of ten-month-old infants to videotaped presentations of happy and sad facial expressions. The authors (measuring EEG activity elicited) found greater left frontal region activity in response to happy than to sad expressions. The authors consider the results to support the existence of a fundamental affective distinction between the hemispheres with regard to approach and avoidance tendencies. This asymmetrical frontal activity may reflect the ability of these facial expressions to elicit approach or avoidance behavior.

Further evidence of the impact of facial expressions on the behavior of infants is provided by Sorce, Emde, Campos, and Klinnert (1985). In this experiment, 12-month-olds were placed in a "visual cliff" situation, a steep grade ending in a four-foot cliff covered by hard, clear plexiglass. Infants were placed on the shallow side of the grade and their mothers produced facial expressions from the cliff side. Mothers smiled at their infants until they reached the edge of the cliff and then changed their expressions. When mothers changed to broader smiles or displayed interest, approximately 75% of the infants crossed the visual cliff toward their mothers. However, when mothers displayed sadness, 33% crossed, during anger, 11% crossed, and during fear expressions none crossed.

Conclusion

Based on the studies reviewed, it seems reasonable to conclude that perceiver reactions to emotionally expressive stimuli are indeed mediated emotionally. Supportive evidence has come primarily from studies of the behavioral reactions of infants and primates, while the results of adult studies have been equivocal due to the lack of clear criteria for the discrimination of emotional and imitative responding. Work in our laboratory is currently in progress which attempts to demonstrate the emotionality of reactions in adult perceivers, employing self-reported and facial response measures. This work employs the additional criterion of response complementarity in inferring emotional mediation.

SUPPLEMENTARITY AND COMPLEMENTARITY

With regard to social communications, the interface of emotions displayed by the sender and receiver may be said to be supplementarily or complementarily related.

Supplementarity

A supplementary emotional reaction is one which is equivalent in type to the emotional expression perceived. Such congruence of emotional quality between model and perceiver has been proposed to form the basis of empathy (Wispe 1986). An example of supplementarity is the well-documented reciprocation of the social smile (Kraut & Johnson 1979). Similarly, empathic responses to expressions of fear (e.g. reactions to horror films), sadness (e.g. reactions to "tear-jerker" films), and pain (e.g. Vaughan & Lanzetta 1980) may be viewed as supplementary.

Complementarity

Complementary interactions are those in which the adaptive emotional response differs from the emotional expression perceived. The prototypical complementary relationship is that of anger and fear. Commentary on this relationship has been surveyed by Plutchik (1980):

"...McDougall (1921) wrote that 'the fear-impulse is the most opposed in tendency to the pugnacious.' Tolman (1923) wrote: 'fear operates by leading away from [and anger] by leading towards and destroying.'...Carr (1929) added (1929) that 'Anger is correlated with an aggressive attack against obstacles, while fear is associated with the opposite type of behavior.'...Cannon (1939) has written about the 'fight or flight' opposition..." (p. 136).

Other potentially complementary relationships between emotions include reactions to disgust and contempt, which along with anger form Izard's (1977) "hostility triad". For example, disgust has been conceptualized as an emotion of fundamental rejection (Tomkins 1982) which, directed socially, is likely to elicit either sadness or anger depending upon the particular relationship in question. Examples of higher-level social affects which are complementary include instances of sadism (schadenfreude) which involves a "perceiver's" joy at another's distress, and envy which involves malice toward another's joy.

Similar Distinctions

The above distinction between complementarity and supplementarity has been made in clinical psychology in describing different modes of psychological "projection" of traits and emotional states. Complementary projection occurs "when an individual projects a trait that is different from his own (and) the projected trait is the complement of his own trait" (Holmes 1968, p. 250). Supplementary projection occurs when "the individual projects onto other people characteristics which are identical to his own.."(Holmes 1968, p. 253).

Similar distinctions have also been proposed by Heider (1958) and by empathy researchers Stotland, Sherman, and Shaver (1971). Stotland et al. (1971) designated instances in which the emotions of the perceiver and the perceived are highly similar as "simple empathy", and instances wherein the dyad experience qualitatively different emotions as "contrast empathy". Heider (1958), in his theory of interpersonal relations, distinguished between "concordant" and "discordant" emotional reactions which are equivalent to simple and contrast empathy, respectively.

Inferring Emotion from Complementarity

A potential means to ascertain whether reactions to expressions of emotion are imitative or emotional would involve the comparison of subject facial reactions with regard to the nature of their correspondence with eliciting stimuli, using complementarily related emotions. As previously indicated, the findings of studies using supplementarily related emotions are inconclusive because such reactions may appear to be the result of either imitation or emotional mediation.

Demonstration of perceiver response complementarity can afford the inference of emotional mediation primarily because such a response is not imitative, by definition. To verify that emotion rather than imitation is the mediator, the response must be found reliably over time, across contexts, and across individuals.

For the response to be labeled emotional it must also meet the following requirements. First, the facial response must conform to the patterns of facial expressions of emotion which are cross-culturally recognized as representative of specific emotions and shown when experiencing the same emotion (unless cultural display rules preclude their exhibition; Ekman and Oster 1982). Second, the response should be adaptive; that is, the response should be a true "complement" to the stimulus expression in terms of its expression of action tendencies (Plutchik 1980). Finally, the response should be validated using multiple methods (i.e. psychophysiological measures, self-reports, and coding of expressive behavior).

EMOTIONAL RELEASERS AND PERSUASION

The finding that emotional expressions can serve as elicitors of specific affective reactions is of fundamental importance to an understanding of the effects of advertising, which now incorporates more emotional expression than ever before (Holman 1986). The psychobiological approach to emotion (Buck 1988; Kroeber-Riel 1979) is an appropriate framework for conceptualizing the eliciting effects of emotional expressions. According to the psychobiological model, the perceptual systems of individuals are biologically "attuned" for the reception of certain classes of stimuli which are important for adaptation (i.e. releasers) and for changing patterns of stimulation. Using these recognition rules, stimuli are "filtered" preconsciously, drawing attention toward biologically important stimuli and away from unimportant or repetitive stimuli. Thus, this model predicts that attention will be drawn toward the emotional expressions of others, a position supported by Kroeber-Riel's (1984) finding of increased frequency of eye fixations with ads containing "emotional pictorial elements", and Lang's (1990) finding that "even a small amount of emotion in a message increases attention to that message" (p. 296). Intimately involved in the operation of the innate releasing mechanism are the attention/activation system (ascending reticular activating system; ARAS) and the subcortical emotional centers (limbic system). There is evidence that the ARAS and limbic structures receive direct inputs from sensory systems, and that these inputs are received before those leading to the neocortex, perhaps establishing a level of arousal and an emotional context for subsequent analytic cognition. The outputs of the attention/activation and emotional systems are communicated to the perceiver as immediate, wholistic sensory experience (syncretic cognition; Tucker 1981) which has been associated with right hemisphere functioning. An implication of the psychobiological model is that automatic operations of the attentional and emotional systems determine to some extent the inputs to analytic cognitive processing and the nature of that processing.

The psychobiological model suggests an important implication for persuasion strategy. Receivers of communications containing emotional releasers are likely to focus on these affective elements, resulting in the elicitation of emotional reactions. The corresponding emotional response (supplementary or complementary) should be experienced before any analytic evaluation, creating a specific emotional context within which to process a neutral stimulus.

A Comparison of Models of Attitude Formation and Change

The features of releaser-based persuasion may be explicated by differentiating such persuasion from the related persuasion mechanisms of peripheral-route processes and direct affect transfer (as in classical conditioning). Whereas the peripheral-route denotes an absence of argument scrutiny, the perception of releasers should increase attention to all message elements, particularly those which are integrated with the releaser through content or other associations (Kroeber-Riel 1979), and should shape interpretation by providing an emotional context. Furthermore, the elicitation of emotion by releasers does not preclude "central" analytic elaboration, which may proceed within the biasing context of the elicited emotion. Thus, releaser-based persuasion is not equivalent to the reliance upon an affective cue under conditions of low elaboration likelihood (cf. Petty, Cacioppo, Sedikides, & Strathman 1988).

Releaser-based persuasion may also be distinguished from the affect transfer of classical conditioning. First, emotional releasers elicit true emotions due to the significance of the information they transmit for the well being of the perceiver. Hence, releasers are not processed in the manner of usual unconditioned stimuli, such as odors and sounds, which produce pleasant or unpleasant sensory experiences. Second, the genuine emotion elicited by releasers should determine the emotional tone of the perceiver's thoughts toward other elements of the communication, providing an emotional grounding for the development of attitudes. As classical conditioning involves the meaningless association of sensations with neutral stimuli, attitudes formed through conditioning are without emotional grounding, possibly resulting in weaker attitude development. Third, the formation of attitudes through exposure to releasers should not necessitate the repeated pairings required for conditioning because of the emotionality experienced at the first exposure (cf. Seligman 1970). This is not to say, however, that repeated pairings would not strengthen the association. The capacity of the association to be strengthened may be particularly relevant in cases wherein an otherwise neutral stimulus can acquire the characteristics of a releasing stimulus, as in package designs featuring releasers (e.g. the smiling Gerber baby, Quaker Oats man, and Aunt Jemima). Finally, while a major concern with classical conditioning is the question of subject awareness of the UCS-CS contingency, whether or not perceivers are aware of the intent behind the use of releasers should not matter, as the process is essentially automatic.

FUTURE RESEARCH

A practical implication of the psychobiological theory discussed herein concerns the methodology of ad testing. As previously indicated, the process of direction of attention may be seen as strongly influenced by subcortical emotional systems. Based on this notion, it is suggested that the elements of ads which are attended to by perceivers will tend to be those which elicit emotion.

While measures of attention such as eye-tracking techniques and recall tests give the researcher a record of the ad information picked-up by perceivers, the motivation behind the direction of attention is untapped. A greatly enriched picture of ad effects would likely emerge with the addition of converging emotional response measures, assessing changes in physiological arousal, expressive behavior, and subjective experience. While arousal may be measured as changes in skin conductance, an economical alternative may be the measurement of eye fixation frequency (Kroeber-Riel 1984). The most promising candidate for the measurement of subtle expressive behavior is facial EMG, which has been demonstrated to differentiate positive and negative reactions (Bagozzi 1991). The insertion of questions regarding the subjective emotional experience of the respondent into post-exposure interviews would serve to complete the measurement of emotional output. Combining these measures would afford the researcher knowledge of which elements were attended to, the valence and quality of perceivers' reactions to those elements, and the intensity of those reactions.

REFERENCES

Bagozzi, R.P. (1991), "The Role of Psychophysiology in Consumer Research," in Handbook of Consumer Behavior, ed. T.S. Robertson and H.H. Kassarjian, Englewood Cliffs, N.J.: Prentice Hall.

Batra, R. and M.L. Ray (1986), "Affective Responses Mediating Acceptance of Advertising," Journal of Consumer Research, 13, 234-249.

Bavelas, J.B., Black, A., Lemery, C.R., and Mullett, J. (1986), "I Show How You Feel: Motor Mimicry as a Communicative Act," Journal of Personality and Social Psychology, 50, 322-329.

Buck, R.(1988), "Nonverbal Communication: Spontaneous and Symbolic Aspects," American Behavioral Scientist, 31, 341-354.

Cohen, J.B. and C.S. Areni (1991), "Affect and Consumer Behavior," in Handbook of Consumer Behavior, ed. T.S. Robertson and H.H. Kassarjian, Englewood Cliffs, N.J.: Prentice Hall.

Darwin, C.R. (1872), The Expression of Emotions in Man and Animals, London: John Murray.

Davidson, R.J. and N.A. Fox (1982), "Asymmetrical Brain Activity Discriminates between Positive and Negative Affecting Stimuli in 10-Month-Old Human Infants," Science, 218, 1235-1236.

Dimberg, U. (1982), "Facial Reactions to Facial Expressions," Psychophysiology, 19, 643-647.

Dimberg, U. (1988), "Facial Electromyography and the Experience of Emotion," Journal of Psychophysiology, 2, 277-282.

Eibl-Eibesfeldt, I. (1989), Human Ethology, New York: Aldine de Gruyter.

Ekman, P. (1982), "Methods for Measuring Facial Action," in Handbook of Methods in Nonverbal Behavior Research, ed. K.R. Scherer and P. Ekman, Cambridge: Cambridge University Press.

Ekman, P. and H. Oster (1982), "Review of Research, 1970-1980," in Emotion in the Human Face, ed. P. Ekman, Cambridge: Cambridge University Press.

Friestad, M. and E. Thorson (1986), "Emotion-Eliciting Advertising: Effect on Long-Term Memory and Judgement," in Advances in Consumer Research, Vol. 13, ed. R. J. Lutz, Provo, UT: Association for Consumer Research.

Heider, F. (1958), The Psychology of Interpersonal Relations, New York: Wiley.

Holman, R.H. (1986), "Advertising and Emotionality," in The Role of Affect in Consumer Behavior: Emerging Theories and Applications, ed. R.A. Peterson, W.D. Hoyer, and W.R. Wilson, Lexington, MA: D.C. Heath and Co.

Holmes, D.S. (1968), "Dimensions of Projection," Psychological Bulletin, 69, 248-268.

Izard, C.E. (1977), Human Emotions, New York: Plenum Press.

Kraut, R.E. and R.E. Johnson (1979), "Social and Emotional Messages of Smiling: An Ethological Approach," Journal of Personality and Social Psychology, 37, 1539-1553.

Kroeber-Riel, W. (1979), "Activation Research: Psychobiological Approaches in Consumer Research," Journal of Consumer Research, 5, 240-250.

Kroeber-Riel, W. (1984), "Effects of Emotional Pictorial Elements in Ads Analyzed by Means of Eye Movement Monitoring," in Advances in Consumer Research, Vol. 11, ed. T. C. Kinnear, Ann Arbor, MI: Association for Consumer Research.

Lang, A. (1990), "Involuntary Attention and Physiological Arousal Evoked by Structural Features and Emotional Content in TV Commercials," Communication Research, 17, 275-299.

Masters, J.C. (1979), "Interpreting 'Imitative' Responses in Early Infancy," Science, 205, 215.

McDougall, W. (1908), An Introduction to Social Psychology, London: Methuen.

McHugo, G.J., J.T. Lanzetta, D.G. Sullivan, R.D. Masters, and B.G. Englis (1985), "Emotional Reactions to a Political Leader's Expressive Displays," Journal of Personality and Social Psychology, 49, 1513-1529.

Petty, R.E., J.T. Cacioppo, C. Sedikides, and A.J. Strathman (1988), "Affect and Persuasion: A Contemporary Perspective," American Behavioral Scientist, 31, 355-371.

Plutchik, R. (1980), Emotion: A Psychoevolutionary Synthesis, New York: Harper & Row.

Sackett, G.P. (1966), "Monkeys Reared in Isolation with Pictures as Visual Input: Evidence for an Innate Releasing Mechanism," Science, 154, 1468-1473.

Schwartz, G.M., C.E. Izard, and S.E. Ansul (1985), "The 5-Month-Old's Ability to Discriminate Facial Expressions of Emotion," Infant Behavior and Development, 8, 65-77.

Seligman, M.E.P. (1970), "On the Generality of the Laws of Learning," Psychological Review, 77, 406-418.

Stotland, E., Sherman, S., and Shaver, K. (1971), Empathy and Birth Order, Lincoln: University of Nebraska Press.

Sorce, J.F., R.N. Emde, J.J. Campos, and M.D. Klinnert (1985), "Maternal Emotional Signalling: Its Effects on the Visual Cliff Behavior of 1-Year-Olds," Developmental Psychology, 21, 195-200.

Tomkins, S.S. (1982), "Affect Theory," in Emotion in the Human Face, ed. P. Ekman, Cambridge: Cambridge Univ. Press.

Tucker, D.M. (1981), "Lateral Brain Function, Emotion, and Conceptualization," Psychological Bulletin, 89, 19-46.

Vaughan, K.B. and J.T. Lanzetta (1980), "Vicarious Instigation and Conditioning of Facial Expressive and Autonomic Responses to a Model's Expressive Display of Pain," Journal of Personality and Social Psychology, 38, 909-923.

Wispe, L. (1986), "The Distinction Between Sympathy and Empathy: To Call Forth a Concept, a Word is Needed," Journal of Personality and Social Psychology, 50, 314-321.

----------------------------------------