Visual discrimination and recognition of facial expressions of anger, fear, and surprise in 4- to 6-month-old infants

Visual Discrimination and Recognition of Facial Expressions of Anger, Fear, and Surprise in 4- to 6-Month-Old Infants JUAN M. SERRANO JAIME IGLESIAS ANGELA LOECHES Departamento de Psicologiu Biolbgica y de la Salud Universidad Autdnoma de Mudrid Madrid, Spain

On the assumption that the ability to discriminate facial expressions has adaptive value to infants during early social exchanges, ethologically based theorists have argued that this ability is innate. Guided by this perspective, we investigated the ability of infants, 4-6 months old to recognize and discriminate facial expressions of anger, fear, and surprise. Results obtained with an infant-controlled habituation-recovery procedure showed that infants both discriminated and recognized these expressions when portrayed by several adult female models. In addition, infants spent more time looking at expressions of anger and surprise than at fear expressions. These results suggest that infants can abstract configurations of features that give affective meaning to facial expressions. It is suggested that the differences in habituation to each expression might be the result of their distinct functional signification for the infant. 0 1992 John Wiley & Sons, Inc.

Over the last 20 years, interest in research on the infants' ability to recognize and discriminate facial expressions of emotions has continued to increase, as shown by the number of recent reviews on this subject (see e.g., Klinnert, Campos, Sorce, Emde, & Svejda, 1983; Nelson, 1985, 1987; Oster, 1981; Walker-Andrews, 1988). Most authors take a psychobiological approach that assumes a Darwinian thesis on the innateness of emotional expression. From an ethological perspective, some theorists suppose that the abilities to express and recognize certain emotions

Reprint requests should be sent to Juan M. Serrano, Departamento de Psicologia Biologics y de la Salud, Facultad de Psicologia, Universidad Autonoma de Madrid, 28049 Madrid, Spain.

Received for publication 19 December 1991 Revised for publication 15 April 1992 Accepted at Wiley 20 April 1992

Developmental Psychobiology 25(6):411-425 (1992) 0 1992 by John Wiley & Sons, Inc. CCC 0012-1630/92/060411-15

412 SERRANO ET AL.

have evolved in parallel as an aid in intraspecific communication and are particu- larly important during the infant’s first social interactions (Ekman, Friesen, & Ellsworth, 1972; Izard, 1971; Tomkins, 1961, 1962).

Although support exists for this perspective, there is still no agreement as to which facial expression infants can recognize and discriminate (see e.g., Ekman & Oster, 1979). Nor is there agreement on which aspects of facial expression guide infant responses. Research on the subject in the first half of this century was characterized by its lack of any measurement or method sensitive to the infant’s perceptive limitations. Based on variations observed in infant gross motor behavior changes, the results from this era led only to the conclusion that recognition and discrimination of expressions of happiness and anger occurred clearly after the 7th month (see e.g., Ahrens, 1954; Biihler & Hetzer, 1928; Spitz & Wolf, 1946).

The limitations of these early studies were significantly reduced in the 1960s, when methods for studying visual preference and habituation-recovery paradigms confirmed that infants’ visual fixation on stimuli was a reliable indicator of their ability to distinguish between different objects (Fantz, 1958; Horowitz, Paden, Bahana, & Self, 1972). Those studies demonstrated that from 3 months on infants can discriminate facial expressions of happiness, anger, fear, surprise, sadness, and disgust shown to them in slides or films (Barrera & Maurer, 1981; Iglesias, 1986; Labarbera, Izard, Vietze, & Parisi, 1976; Wilcox & Clayton, 1968; Young- Browne, Rosenfeld, & Horowitz, 1977).

The common characteristic of these studies was that the different facial expressions shown were simulated by a single model. It could then be insured that variations in visual response to each facial expression were due to the perception of changes in the inner facial features, because it was the only physical feature that varied from expression to expression. However, it was not possible to determine whether the perceived changes were the configuration of features relevant that composed each expression, or some isolated physical features of the model’s face, having no emotional value. Researchers also could not be certain that infants distinguished facial expressions as different emotional categories, based upon the abstraction of invariant facial characteristics (Nelson, 1987; Walker-Andrews, 1988). This categorial perception is on the basis of adult recognition and discrimination processes, making it necessary to discover whether infants are able to perceive facial expressions in a similar fashion.

This question has been studied by using facial expressions simulated by different models as stimuli. In this case, the infant must first perceive the similarity in the facial expression of a series of different models, and then a) generalize recognition to new examples of the same expressive category, and b) discriminate facial expressions which do not belong to the same category. Several researchers have used this strategy to show that 4- to 9-month-old infants perceive facial expressions of happiness and perhaps others such as fear, sadness, or surprise (Caron, Caron, & MacLean, 1988; Caron, Caron, & Myers, 1982, 1985; Nelson, 1987; Ludemann & Nelson, 1988; Nelson & Dolgin, 1985; Nelson, Morse, & Leavitt, 1979). Some expressions, however, turned out to be more easily distinguishable than others, depending on the order in which they were displayed. Thus, for instance, Nelson and colleagues (1979), Nelson and Dolgin (1985), and Ludemann and Nelson (1988) have shown that fear is differentiated from happiness when the infant first has been habituated to happiness expressions, but not vice versa.

RECOGNITION OF FACIAL EXPRESSION IN INFANTS 413

Another underlying question that also remains is at what age infants can recognize and discriminate some facial expressions of emotions. Research using only one model shows that infants can discriminate certain facial expressions at 3 months of age. However, other studies employing several models as stimuli suggest that this capacity emerges anywhere from 4-9 months (Caron et al., 1982, 1985; Ludemann & Nelson, 1988; Nelson, 1987; Nelson & Dolgin, 1985). It is possible that the discrimination of expressions emerges at different ages, but almost no work done with several models has studied infants less than 6 or 7 months old.

Beginning at about 2 months, the infant’s visual acuity and contrast sensitivity permits the most important features of the face to be perceived, at least at close distance (Atkinson, 1984; Banks & Salapatek, 1983). At this time the infants also begin to scan the inner face extensively, and show an initial preference for the eyes (Haith, Bergman, & Moore, 1977; Maurer, 1985). Otherwise, the ability to abstract the invariant characteristics of stimuli seems to be inherent to the visual system; very simple configurations have been shown to be perceived in this way even by the newborn (Antell, Caron & Myers, 1985; Dodwell, Humphrey, & Muir, 1987). So, infants under 6 months visually scan faces in a manner consistent with being able to recognize facial expressions, and also prior to 6 months, infants’ acuity and contrast sensitivity is such that they should have little trouble detecting the high contrast features that denote facial expressions.

Another major aspect to consider is the validation and reliability of the stimuli used. In most cases, slides of facial expressions were used instead of films or live models in order to guarantee that variations in the infant’s visual fixation could be attributed solely to the expressions and not to factors such as voice or movement. However, only some of these studies attempted to ensure that the stimuli shown were reliable prototypes of each emotion. Some used inexperienced people to simulate expressions, such as the infants’ own mothers, and did not evalute the similarity of the different samples of the same emotion.

From our point of view, the stimuli should be selected by more objective criteria, such as interobserver agreement (see e.g., Ludemann, 1991; Ludemann & Nelson, 1988; Nelson & Dolgin, 1985) and the analysis of expressive movements with facial activity measurement codes (Iglesias, 1986). The selection of reliable stimuli is especially important in studies that use several models to simulate the same emotion. The infant’s ability to perceive facial expressions categorically, based on an abstraction of the characteristics common to different people, cannot be tested when the pertinent features are not constant. Furthermore, Ludemann (1991) recently made it clear that, at least until 7 months of age, the presence of specific prototypical features is fundamental to the discrimination of facial expressions.

Finally, previous studies also have focused primarily on the infants’ visual response to happy faces and paid less attention to their response to other emotional faces. Some psychobiological theories of emotion assume the existence of early recognition of the facial expressions of other basic emotions such as fear, anger, surprise, sadness, and disgust (see e.g., Ekman et al., 1972; Izard, 1971). So, it is important to objectively show that the infant recognizes and differentiates these expressions.

The present study, guided by an ethological perspective, investigated the ability of infants 4-6 months of age to recognize and discriminate between facial

414 SERRANO ET AL.

expressions of the emotions of anger, fear, and surprise. It is important to stress that no previous work done with several models has studied the infant’s perception of the expressions of anger and fear together. This comparison provided the opportunity to observe the infant’s response to two negative emotional expressions, similar in their configurational complexity, but with a clearly differentiated communicative value or meaning.

Fear and surprise expressions look alike. Both expressions are characterized by the raising of the eyebrows and the opening of the mouth, displaying the teeth. Furthermore, the emotional significance of the two expressions also is in part similar. In fact, fear and surprise are the facial expressions that adults of various cultures find the most difficult to distinguish (Ekman, Sorenson, & Friesen, 1969). So, it is of interest to determine if the infant can discriminate between the expressions of fear and surprise.

Method

Subjects The subjects were 12 infants (6 males, 6 females) 4-6 months of age (range =

18-27 weeks; X = 22 weeks). All were full term (37-42 weeks of gestation), were born weighing 2.6 to 3.9 kg, and showed no perception or motor irregularities according to pediatric examinations prior to the study.

The planned tests could not be completed on five cases, either because the infants cried during testing (n = 3) or paid no attention to the stimuli ( n = 2). According to their parents or caregivers, these interruptions were justified by recent slight illnesses or vaccinations on the day of or the days immediately preceding the tests. These infants were tested again another day.

Stimuli

The stimuli were color slides of six females depicting facial expressions of anger, fear, or surprise. These slides were a selected sample of stimuli from 20 female models that were tested in a pilot study. Before simulating the test expressions, the models observed samples of angry, fearful, and surprised faces taken from the Ekman and Friesen (1975) study, and were required to practice the facial movements defining these emotions in front of a mirror.

The initial series of slides was shown to a group of 38 psychology undergradu- ate students, who judged them according to a closed list of six emotions (happiness, anger, fear, surprise, disgust, and sadness) and a neutral category. The five facial expressions recognized by the greatest number of judges were selected for use in this study. These faces are presented in Figure 1. The percentage of judges who correctly identified each expression always exceeded 75%.

Lastly, all selected facial expressions were further analyzed using the Facial Action Coding System (FACS) designed by Ekman and Friesen (1978) to insure that all faces matched one of the prototypes of anger, fear, and surprise these authors considered universal.


ANGER FEAR SURPRISE

Fig. I . The facial expressions of anger, fear, and surprise used as the stimuli.

416 SERRANO ET AL.

Apparatus and Procedure The slides were shown with a Zeissikon projector on a 75 x 65 cm retroprojec-

tion screen. The screen was equipped with two 4-cm viewing slots on each side to enable two observers to independently observe the corneal reflection of the stimulus on the infant’s pupil (see e.g., Maurer, 1975). One observer controlled the slide projector while the other was unaware of the stimulus being provided. The infant was seated 30 to 40 cm from the screen in a chair designed especially to secure the infant by the waist without impeding head and limb movement. Testing was generally done after feeding and outside normal sleep periods.

Facial expressions were presented on the screen similar in size to an adult face (approx. 23 x 18 cm). In Phase 1, the habituation phase, the infant was shown slides of three different models expressing the same emotion. Only one slide appeared and remained on the screen until the subject ceased looking at it for at least 2 s. The next trial followed after a 4-s interval. This sequence was repeated until the infant’s visual fixation response decreased to a preestablished habituation criterion. To meet criterion, the time the infant fixated on the stimulus on the last three trials could not exceed 50% of the average of fixation on the first three trials. When criterion was reached, phase 2 was initiated. The infant was shown slides, one at a time, of two new models expressing 1) the emotion to which the infant had been habituated, and 2) a new emotion. The order in which the two new stimuli were presented was counterbalanced.

A within-subject experimental design was used. Each infant was subjected to the six experimental conditions resulting from combining the three expressions shown two at a time (habituated expressionhew expression: angerlfear, fear1 anger, angerlsurprise, surpriselanger, fearlsurprise, surprisejfear). The order of exposure of each subject to each experimental condition was randomized and therefore variable from one subject to another, with a minimum interval of 3 days between consecutive tests with the same infant (Rovee-Collier, 1987).

Results

Reliability The interobserver reliability level calculated using the Pearson product-

moment coefficient was 0.93 and ranged between 0.80-0.99. Given the high interobserver reliability, the data used in statistical analyses were the average of the visual fixation times registered by both observers on each trial.

Habituation Phase

Two complementary measurements of the infant response to each expression were considered, cumulative fixation time and number of trials to reach criterion. In both cases, the data were evaluated using a mixed model ANOVA that considered the within-subject effects of the expression factor and between-subject effects of the sex factor [expression(3) x sex(2)I.

The analyses revealed significant differences between the cumulative visual fixation time for the different expressions during the habituation phase, F(2, 33)


; 100

; 90

80 -

n 7 0 -

s 60 50

40

d 30 - s

20

10

0

130

.. 110

--

Anger Fear Surprise

Fig. 2. The cumulative visual fixation times to each of the facial expressions during the habituation phase of the experiment.

= 3.79; p < .05, but not in the number of trials needed in each case. Also, no significant differences were due to sex or an interaction between sex factor and the facial expression.

In order to find the reason for the significant effect of the expression factor, each expression’s cumulative fixation time was compared to the rest using the Dunn-Bonferroni t test. The analyses showed that the fear expression was viewed significantly less than the anger and surprise expressions, t ( 3 , 3 3 ) = 2.53; p < .05. Therefore, though the infants became habituated to the three expressions in the same number of trials, they paid more attention to the expressions of anger and surprise than fear (see Fig. 2).

Test Phase In order to rule out the possible influence of the novelty of the test phase

models, the visual fixation time on each of the two new models shown in this phase was contrasted. The order of exposure of these models meant that the first model first depicted the new expression and then the habituated one, while the second model followed the opposite order. Thus, when the first model simulated the new expression, she showed the infant both a new expression and a new physiognomy; however, the second model showed only a new facial expression. Because of her different degree of novelty, the first model might receive more visual fixation time than the second in the case of a new expression, while the second model would be viewed more when making the habituated expression.

Twelve repeated measures ANOVAs were conducted, two per experimental condition, aimed at comparing the amount of visual fixation time on the first and second models, both when they simulated the new expression and when they modeled the habituated expression. These analyses revealed no significant differences, ps > .05 in all cases. That is, neither the first model simulating the new expression nor the second model simulating the habituated expression received more visual fixation than the other.

418 SERRANO ET AL.

The data were further analyzed to determine if the infants discriminated between the new expression and the habituated expression and whether they recognized the new examples of the habituated expression. Six mixed model ANOVAs were conducted, one per experimental condition, taking into account the possible within subject effect of the expression and the between-subject effect of sex. The three levels of the expression factor were: the average visual fixation time during the last three trials of the habituation phase, the average fixation time for the habituated expression in the test phase, and the average visual fixation time for the new expression.

As shown in Table 1, the results of these six analyses were essentially similar, denoting the statistical significance of the effects due to the expression factor. However, in no case were significant differences observed between the visual fixation of male and female infants, nor was there interaction between the expression and the sex factor.

Given the statistical significance of the expression factor, Dunn-Bonferroni t tests were done for each experimental condition in order to contrast, on the one hand, the average visual fixation times given to the new expression as compared to the habituated expression, and, on the other hand, the average visual fixation times registered for the habituated expression in the last trials of the habituation phase with regard to those registered during the test phase.

In all experimental conditions, the infants looked more at the new expression than the habituated expression in the last trials of the habituation phase, p < .01, see Table 2. Significant differences, p < .01, were also obtained between the fixation times on the new and habituated expressions in the test phase, except in the surprise/fear condition. That is, the infants demonstrated that they could discriminate between the three expressions under all conditions except this one. Nevertheless, the mean visual fixation times for the surpriseifear condition followed the same tendency as observed in the other experimental conditions (see Table 2 and Fig. 3).

Moreover, under five of the six experimental conditions, the infants looked at the habituated expression for the same amount of time in the habituation phase and the test phase. Significant, p < .05, see Table 2, differences were obtained between the two levels of the habituated expression in one of the two times at which the infants were habituated to fear (fear/anger condition). However, it must be pointed out that the average visual fixation times in the fear/anger condition also were similar to those observed in the other experimental conditions, suggesting the recognition of the expressive similarity (see Fig. 3).

Discussion The present results show that infants can discriminate between the new and

the habituated facial expressions under five of the six experimental conditions studied. Most importantly, they identified the habituated expression even though it was simulated by different faces, again with the exception of one experimental condition. Thus, we may infer that 4- to 6-month-old infants recognize the facial expressions of anger, fear, and surprise. However, before discussing the implica- tions of this interpretation, other alternative explanations of these results will be considered.

Tabl

e I

Res

ults

of

AN

OV

As

Des

igne

d fo

r A

sses

sing

the

Eff

ects

of E

xpre

ssio

n an

d Se

x F

acto

rs in t

he T

est P

hase

E

xper

imen

tal C

ondi

tion

Ang

er/F

ear

Fear

/ Ang

er

Ang

edSu

rpri

se

Surp

rise

/ Ang

er

Fead

Surp

rise

Su

rpri

se/F

ear

Eff

ect

df

S.S

. F

p S

.S.

F

p S

.S.

F p

S.S.

F

p S

.S

.F

p

S.

S.

F

p -.

Exp

ress

ion

2 22

9.7

7.5

.W** 1

57.6

18

.8 .W** 35

6.2

10.4

.O** 15

6.6

8.4

.W**

70

.0

9.3

.OO*

* 71

.7

4.9

.01*

Se

x 1

22.7

.7

.4

2 28

.2

1.9

.I9

12.3

.4

.5

3 3.

4 .I

.7

1 3.

4 .4

.5

6 12

.8

.4

.56

Exp

r. x

Sex

2

32.8

1.

1 .3

6 22

.4

2.5

.I0

13.1

.3

.6

9 .7

.I

.%

9.

5 1.

3 .3

0 24

.3

1.7

.21

* p <

.05.

**

p <

.01.

a

%

420 SERRANO ET AL.

Table 2 Results of Dunn-Bonferroni T Test Comparing Visual Fixation Times to Last Three Habituation Phase Trials, Habituated Expression in Test Phase, and to New Expression

Difference tn P

Anger/Fear Anger (habituation)/Fear Anger (test)/Fear Anger (habituation)/Anger (test)

Fear (habituation)/Anger Fear (test)/Anger Fear (habituation)/Fear (test)

Anger (habituati0n)iSurprise Anger (test)/Surprise Anger (habituation)/Anger (test)

Surprise (habituation)/Anger Surprise (test)/Anger Surprise (habituation)/Surprise (test)

Fear (habituation)/Surprise Fear (test)/Surprise Fear (habituation)/Fear (test)

Surprise (habituation)/Fear Surprise (test)/Fear Surprise (habituati0n)Surprise (test)

Fear/ Anger

Anger/Surprise

Surprise/Anger

Fear/Surprise

Surprise/Fear

5.89 > 3.94 < .01 4.58 > 3.94 < .01 1.78 < 3.14 > .05*

7.08 > 2.11 < .01 3.16 > 2.11 < .01 1.90 > 1.68 < .05

7.24 > 4.22 < .01 6.15 > 4.22 < .01 1.09 < 3.37 > .05*

5.05 > 3.27 < .OI 3.25 > 3.27 < .01 1.30 < 2.44 > .05*

2.92 > 1.95 < .01 3.00 > 1.95 < .01

.85 < 1.56 > .05*

3.46 > 2.73 < .01 1.64 < 2.18 > .05* 1.81 < 2.18 > .05*

* ns.

As indicated above, the first factor that had to be ruled out was the possible influence of the novelty of the models shown during the test phase. This influence did not determine the registered visual fixation times. Although infants displayed a slight tendency to look more at the new examples of the habituated expression, this difference was significant on only one occasion. Furthermore, in general, the infants looked significantly more at the new expression than the habituated one though both were portrayed by new faces. Moreover, there was no variation in the visual fixation times given to each of the two models shown during this test phase despite their different novelty, both when they simulated the new expression and when they displayed the habituated expression. Consequently, it is improbable that the novelty of the face determined the marked ability for discrimination and recognition of facial expressions observed in this study. It is even possible that the small effect of novelty would have been totally eliminated with the presentation of a larger number of models during the habituation phase (Fagan, 1976; Nelson et al., 1979). Caron and colleagues (1982, 1985) demonstrated that 4-month-old infants are increasingly able to generalize the recognition of expressions of happiness or anger to new faces as the number of models depicting the expressions in the first habituation phase increases.


= L e a l hmbllue Ir lalm mi3Hmbll * i p r In I a a l = N e w mrpr-mslon

I - __ - . . - - . __

9 --

7 a - - m fl 7 --

i n 6 -

: 5 -

Angmr1Fm.r FsmrlAngar Angar lOurp. Burp.lAngmr F m a r l S u r P . Surp. lFbar

Experimental condition Fig. 3. Visual fixation time to the habituated expression during the last phase of habituation and

in the test phase, and visual fixation time to the new expression presented in the test phase.

It is also possible, however, that the differences observed in visual fixation times to the new expression in comparison with the habituated expression were due only to the fact that the infant prefers to look at the physical characteristics of the face which lack any emotional significance. To be more specific, Caron and colleagues (1985) and Oster and Ewy (1980, cited in Oster, 1981) have indicated that the brightness contrast between the teeth and the lips (the toothiness) is the fundamental characteristic used by 4- and 5-month-old infants to discriminate between facial expressions of happiness and anger or happiness and sadness.

On this possibility, it must be pointed out that in our study the models were not specifically instructed to display certain contrasts or different intensities of expression. However, Figure 1 shows that the models did not simulate the same expression with the same amount of openness of mouth and eyes; and therefore toothiness contrast was variable. Since these physical characteristics changed with different examples of the same expression, it would be difficult for the infants to perceive them as a common attribute to which they could become habituated. In reality, the only thing that did not change from one model to another during the habituation phase was the expressive attributes. Consequently, it is more probable that the infants became habituated to these attributes.

Along these same lines, Nelson (1987) has suggested that the infant’s preference for certain brightness contrasts does not necessarily indicate an inability to perceive the expressive attributes that define emotion, but rather that the effects of these contrasts sometimes occlude the real discriminative abilities of infants. In fact, when this effect is eliminated or properly controlled, the features forming expressions of happiness can be perceived and discriminated from expressions of

422 SERRANO ET AL.

fear, surprise, and anger (see e.g., Kestenbaum & Nelson, 1990; Ludemann & Nelson, 1988).

It is difficult to explain why infants that were habituated to the surprise expression did not discriminate between the expressions of surprise and fear during the test. Ludemann and Nelson (1988), in fact, found just the opposite result. They reported that infants 7 months old discriminated between fear and surprise if they had been habituated to the surprise expression but not if they had been habituated to the fear expression. They suggested that infants fail to make this discrimination because they had less preexperimental experience with the fear than with the surprise expression. Even though they had been habituated to the fear expression, this differential experience would lead them to fixate the fear more than the surprise expression in the test phase.

Yet, our results do not unequivocally support this explanation of why infants failed to discriminate between surprise and fear when surprise was the habituated expression. If one considers cumulative visual fixation times during the habituation phase, it is apparent that fear was the expression which received the least visual fixation.

Nevertheless, our data do show a certain agreement with those found by Ludemann and Nelson (1988). Having ruled out the influences of facial novelty and physical contrasts, we may suppose that the infant prefers to pay attention to the configurational features that provide facial expressions with emotional significance. We may suppose that, during the habituation phase, the infant abstracts the expressive characteristics common to a series of faces in order to be able later, in the test phase, to identify them or distinguish them in different models. This conclusion is also consistent with the other results with 4- and 5-month-old infants, in which generalized recognition and discrimination of facial expressions of happiness and surprise and happiness and fear, respectively, were observed under certain conditions (see e.g., Caron et al., 1982; Nelson, 1987).

Granted that the infant attends to pertinent expressive features and distin- guishes facial expressions as independent categories, we may also believe that infants are capable of identifying the underlying emotional message in each of these expressions (see e.g., Klinnert et al., 1983; Nelson, 1987). Therefore, it would be interesting to discuss our results suppossing that the ability to identify the emotional significance of certain facial expressions is innate, that is, does not depend on the infant’s postnatal experience.

From this perspective, it seems logical that the infant attends most to facial expressions that have the most adaptative value. Consequently, that the infants fixated more on the expressions of anger and surprise than on the fear expressions during habituation may reflect a difference in the importance of the message they transmit to the infant. Thus, it is possible that the alert signal entailed by the expression of surprise is markedly useful from the first few months of life for the regulation of the first social exchanges. Recognition of the expression of surprise might be important because that expression signals the occurrence of other significant events. In fact, it frequently serves to prepare the communicative exchanges that occur immediately thereafter (Eibl-Eibesfeldt, 1989). Mothers usually use this expression to initiate or vary the direction of communication with their infants, and concordance between mother-infant behavior is high when she expresses surprise (Malatesta, Culver, Tesman, & Shepard, 1989).


Moreover, it is probable that the perception of the interaction-interrupting message entailed in the expression of anger is crucial for infants. It also has been suggested that anger displays may function as a discrete social signal from a very early age (Stenberg & Campos, 1990). Angry faces are observed frequently in natural mother-infant interaction conditions, with the purpose of displaying the mother’s disapproval of the infant’s behavior and brusquely interrupting communi- cations (Malatesta & Haviland, 1982). In favor of the early recognition of anger, Sackett (1966) demonstrated that from approximately the first 2 months of age, rhesus monkeys reared in isolation displayed disturbed behavior in reaction to slides of threatening facial expressions by adult rhesus monkeys. This behavior did not appear in reaction to other slides depicting scenes of play or feeding, leading Sackett to consider that the perception of the threat message associated with angry expressions had a genetic origin.

Unlike surprise and anger, the expression of fear may not be so relevant for the infant until a later age. Fear expression appears in ontogeny at about the same time the infant begins to locomote and thereby can be exposed to a wide variety of potentially dangerous situations (Campos, Hiatt, Ramsay, Henderson, & Svejda, 1978; Freedman, 1974). Furthermore, the expression of fear in mother-infant interactive sequences rarely occurs during the first months of life (Malatesta & Haviland, 1982). Infants do not display the facial expression of fear until they are about 6 months old, when approached by a stranger (Iglesias, Loeches, & Serrano, submitted; Izard, Huebner, Risser, McGuinnes, & Dougherty, 1980).

This interpretation of our results implies a biological predisposition for recog- nizing certain facial expressions, but does not preclude the influence of learning. From our point of view, the gradual learning of the instances in which a facial expression appears and the various shades of meaning it can have are based on an initial ability to perceive a basic emotional meaning. We suggest that this innate capacity is useful only in a very limited fashion, as it is restricted to the establish- ment of the first contacts between the infant and the people in hidher immediate surroundings (see also Oster, 1981; Saarni, 1979). The study of other measures of infant behavior, as well as their variations far from the first year of life could be useful ways to know the true abilities and the development of the recognition of emotion facial expressions.

Notes Preparation of this manuscript was supported by a grant from the Direccion General de Investigo-

cidn Cienf$ca y Te‘cnica (PB86-0116). Portions of these data were presented at the 19th Annual Meeting of the British Psychophysiologi-

cal Society and the Annual Scientific Meeting of the Psychobiology Section of the British Psychological Society (Egham, UK, 1991).

Reprint requests should be sent to the first author at the Departamento de Psicologia Biologica y de la Salud, Facultad de Psicologia, Universidad Autonoma de Madrid, 28049-Madrid, Spain.

We would like to thank the anonymous reviewers for their helpful comments.

References Ahrens, R. (1954). Beitrag zur Entwicklund des Physiognomie und Mimikerkennens. Zeitschrit fiir

Antell, S . E., Caron, A. J., & Myers, R. S. (1985). Perception of relational invariants by newborns. Experimentelle und Angewandte Psychologie, 2, 412-454, 599-633.

Child Development, 54, 695-701.

424 SERRANO ET AL.

Atkinson, J. (1984). Human visual development over the first months of life: A review and a hypothesis. Human Neurobiology, 3, 64-74.

Banks, M. S. , & Salapatek, P. (1983). Infant visual perception. In M. M. Haith & J . J . Campos (Eds.), Handbook of child psychology: Vol. 2. Infancy and developmental psychobiology (pp. 435-572). Chichester: Wiley.

Barrera, M. E., & Maurer, D. (1981). The perception of facial expressions by three month olds. Child Development, 52, 203-206.

Buhler, C., & Hetzer, H. (1928). Das erste Verstandis fur Ausdruck im ersten Lebensjahr. Zeitschrift fur Psychologie, 107, 50-61.

Campos, J., Hiatt, S., Ramsay, D., Henderson, C., & Svejda, M. (1978). The emergence of fear of heights. In M. Lewis & L. A. Rosenblum (Eds.), The development of affect (pp. 149-182). London: Plenum Press.

Caron, R. F., Caron, A. J., & MacLean, D. J. (1988). Infant discrimination of naturalistic emotional expressions: The role of face and voice. Child Development, 59, 604-616.

Caron, R. F., Caron, A. J., &Myers, R. S. (1982). Abstraction of invariant face expressions in infancy. Child Development, 53, 1008-1015.

Caron, R. F., Caron, A. J., & Myers, R. S . (1985). Do infants see emotional expressions in static faces? Child Development, 56, 1552-1560.

Dodwell, P. C., Humphrey, G. K., & Muir, D. W. (1987). Shape and pattern perception. In P. Salapatek & L. B. Cohen (Eds.), Handbook of infant perception: Vol. 2 . From perception to cognition (pp. 1-77). London: Academic Press.

Eibl-Eibesfeldt, I. (1989). Human ethology. Hawthorne, NY: Aldine de Gruyter. Ekman, P., & Friesen, W. (1975). Unmasking the face. Englewood Cliffs, NJ: Prentice-Hall. Ekman, P., & Friesen, W. (1978). Facial action coding system. Palo Alto, CA: Consulting Psychologist. Ekman, P., Friesen, W., & Ellsworth, P. (1972). Emotion in the humanface. New York: Pergamon. Ekman, P., & Oster, H. (1979). Facial expressions of emotion. Annual Review of Psychology, 30,

Ekman, P., Sorenson, E. R., & Friesen, W. (1969). Pan-cultural elements in facial expressions of

Fagan, J. F. (1976). Infants’ recognition of the invariant features of faces. Child Development, 47,

Fantz, J. F. (1958). Pattern vision in young infants. Psychological Record, 8, 43-49. Freedman, D. C. (1974). Human infancy: An evolutionary perspective. Hillsdale, NJ: LEA. Haith, M. M., Bergman, T., & Moore, M. J. (1977). Eye contact and face scanning in early infancy.

Horowitz, F. D., Paden, L., Bahana, K., & Self, P. (1972). An infant control procedure for studying

Iglesias, J. (1986). Expresibn facial y reconocimiento de emociones en la infancia. Unpublished doctoral

lglesias, J., Loeches, A., & Serrano, J. M. Submitted for publication. Facial expression of basic

Izard, C. E. (1971). The face of emotion. London: Plenum Press. Izard, C. E., Huebner, R. R., Risser, D., McGuinnes, G. C., & Dougherty, L. M. (1980). The young

infant’s ability to produce discrete emotion expressions. Developmental Psychology, 19,418-426. Kestenbaum, R., & Nelson, C. (1990). The recognition and categorization of upright and inverted

emotional expressions by 7-month-old infants. Infant Behavior and Development, 13, 497-51 I . Klinnert, M. D., Campos, J. J., Sorce, J. F., Emde, R. N., & Svejda, M. (1983). Emotions as behavioral

regulators: Social referencing in infancy. In R. Plutchik & H. Kellerman (Eds.), Emotion: Theory, research, and experience (Vol. 2 ) : Emotions in early development (pp. 57-86). London: Academic Press.

L.abarbera, J. D., Izard, C. E., Vietze, P., & Parisi, S. A. (1976). Four- and six-month-old infants’ visual responses to joy, anger, and neutral expressions. Child Development, 47, 535-538.

L%udemann, P. (1991). Generalized discrimination of positive expressions by seven- and ten-month-old infants. Child Development, 62, 55-67.

Ludemann, P., & Nelson, C . (1 988). Categorical representation of facial expressions by 7-month-old infants. Developmental Psychology, 24, 492-501.

Malatesta, C. Z., & Haviland, J. M. (1982). Learning display rules: The socialization of emotion expression in infancy. Child Development, 53, 991-1003.

527-554.

emotion. Science, 164, 86-88.

627-638.

Science, 198, 853-855.

infant visual fixations. Developmental Psychology, 7 , 90-96.

dissertation. Universidad Autbnoma de Madrid, Madrid.

emotions in infancy.


Malatesta, C. Z., Culver, C., Tesman, J. R., & Shepard, B. (1989). The development of emotion expression during the first two years of life. Monographs of the Society for Research in Child Developmenr, 54, 1-104.

Maurer, D. (1975). Infant visual perception: Methods of study. In L. B. Cohen & P. Salapatek (Eds.), Infant perception: From sensation to cognition: Vol. I . Basic visualprocesses (pp. 1-76). London: Academic Press.

Maurer, D. (1985). Infants’ perception of facedness. In T. M. Field & N. A. Fox (Eds.), Social perception in infants (pp. 73-100). Norwood, NJ: Ablex.

Nelson, C. A. (1985). The perception and recognition of facial expressions in infancy. In T. M. Field & N. A. Fox (Eds.), Social perception in infants (pp. 101-125). Norwood, NJ: Ablex.

Nelson, C. A. (1987). The recognition of facial expressions in the first two years of life: Mechanisms of development. Child Development, 58, 889-909.

Nelson, C. A,, & Dolgin, R. (1985). The generalized discrimination of facial expressions by seven- month-old infants. Child Development, 56, 58-61.

Nelson, C. A., Morse, P. A., & Leavitt, L. A. (1979). Recognition of facial expressions by seven- month-old infants. Child Development, 50, 1239-1242.

Oster, H. (1981). Recognition of facial expressions in infancy?. In M. E. Lamb & L. Sherrod (Eds.), Infant social cognition: Empirical and theoretical considerations (pp. 85-125). Norwood, NJ: Ablex.

Oster, H., & Ewy, R. (1980). Discrimination of sad vs. happy faces by 4-month-old infants: When is a smile seen as a smile?. Unpublished manuscript. University of Pennsylvania, Philadelphia.

Rovee-Collier, C. (1987). Learning and memory in infancy. In J. D. Osofsky (Ed.), Handbook of infant development (2nd ed . ) (pp. 98-148). Chichester: Wiley.

Saarni, C. (1979). Children’s understanding of display rules for expressive behavior. Developmental

Sackett, G. P. (1966). Monkeys reared in isolation with photographs as visual stimuli. Science, 154,

Spitz, R., & Wolf, K. M. (1946). The smiiing response: A contribution to the ontogenesis of social relations. Genetic Psychology Monographs, 34, 57-125.

Stenberg, C. R., & Campos, J. J. (1990). The development of anger expression in infancy. In N. L. Stein, B. Leventhal, & T. Trabasso (Eds.), Psychological and biological approaches to emotion (pp. 247-282). Hillsdale, NJ: LEA.

Tomkins, S. S. (1961). Affect, imagery and consciousness. Vol. 1 . The positive affects. New York: Springer-Verlag.

Tomkins, S. S. (1962). Affect, imagery and consciousness: Vol. 1 . The negative affects. New York: Springer-Verlag.

Walker-Andrews, A. (1988). Infants’ perception of the affordances of expressive behaviors. In C. Rovee-Collier & L. P. Lipsitt (Eds.), Advances in infancy research (Vol. 5 ) (pp. 173-221). Norwood, NJ, Ablex.

Wilcox, B. M., & Clayton, F. L. (1968). Infant visual fixation on motion pictures of the human face. Journal of Experimental Child Psychology, 6 , 22-32.

Young-Browne, G., Rosenfeld, H. M., & Horowitz, F. D. (1977). Infant discrimination of facial expressions. Child Development, 48, 555-567.

Psychology, 15, 424-429.

1468- 1473.

Visual discrimination and recognition of facial expressions of anger, fear, and surprise in 4- to 6-month-old infants

Documents