Attention-orienting and attention-holding effects of faces on 4- to 8-month-old infants

Abstract

Attention-orienting and attention-holding effects of faces were investigated in a sample of 64 children, aged 4 to 8 months old. A visual preference task was used, in which pairs of faces and toys were presented in eight 10-second trials. Effects of age and sitting-ability were examined. Attention-orienting toward faces was measured using the direction of infants’ first looks toward faces. The effect of attention-holding of faces was measured by calculating infants’ face preference scores at 1-second time intervals across the duration of each trial. Faces were not found to attract infants’ first looks significantly more than chance. However, during the first second of looking-time, infants displayed a face preference that was maintained throughout trial length. This attention-holding effect by faces was not related to sitting-ability or age.

Keywords

attention face preference infant sitting ability

Infants’ attention toward faces has been of interest to researchers for decades. A number of studies, dating back to Fantz’ classic work using the first visual preference test (1961, 1963), indicate that infants, even neonates, are more interested in faces or face-like stimuli than other visual stimuli (Easterbrook, Kisilevsky, Muir, & Laplante, 1999; Frank, Vul, & Johnson, 2009; Goren, Sarty, & Wu, 1975; Johnson, Dziurawiec, Ellis, & Morton, 1991; Morton & Johnson, 1991; Valenza, Simion, Cassia, & Umiltà, 1996).

Most of the studies previously conducted on infants’ face preferences have measured how long a face stimulus holds infants’ attention. Cohen (1972, 1976) posited that attending to any visual stimulus involves at least two attentional processes: (1) the attention-getting mechanism, which involves infants’ orienting to a stimulus (henceforth referred to as “attention orienting” to be consistent with recent literature), and (2) the attention-holding mechanism, which involves infants’ fixating on a stimulus. To understand the mechanisms that underlie infants’ face preference, an important question to address is whether faces affect both of these attentional processes.

Two studies that have addressed the effects of faces on the attention-orienting and attention-holding processes used a task in which a face was presented in an array of objects, with infants 3–6 months of age. Gliga, Elsabbagh, Andravizou, and Johnson (2009) recently demonstrated that when 6-month-old infants were presented a color array consisting of a face and five common objects, infants committed a larger proportion of first looks to the face stimulus (that is, attention-orienting). Di Giorgio, Turati, Altoè, and Simion (2012) found developmental differences between 3- and 6-month-olds using a black and white array consisting of a face and either three or five common objects. In contrast to the attention-orienting effect demonstrated in Gliga et al. (2009), Di Giorgio et al. found no evidence that faces affected the direction of first looks at either age. They did, however, demonstrate a developmental shift in the attention-holding effect of faces between 3 and 6 months.

A third recent study, using a preferential looking task, found a developmental difference in how faces affect both the attention-orienting and attention-holding processes between 3 and 5 months of age. Libertus and Needham (2011) found that when 5-month-olds were presented with a face adjacent to a colorful toy, infants demonstrated an initial orienting preference as well as a greater proportion of looking-time toward faces. In contrast, 3-month-olds showed no orienting or looking-time preference for faces. A second group of 3-month-old infants, who received experience manually manipulating objects with “sticky mittens,” behaved like 5-month-olds. These infants oriented more toward faces and their attention was held longer by faces than objects.

Thus far, studies that have investigated both attention-orienting and attention-holding effects of faces consistently show that infants have a face preference (that is, their attention is more effectively held by faces than objects) at around 5–6 months of age, but not at 3 months. However, the effect of faces on infants’ attention orienting is mixed, with two studies showing that attention orienting is affected by faces around 5–6 months (Gliga et al., 2009; Libertus & Needham, 2011), and one study showing no evidence of an effect on attention orienting in 3- or 6-month-olds (Di Giorgio et al., 2012). One possible explanation for this pattern of results is that the effect of faces on holding infants’ attention is robust around 5–6 months, whereas the effect of faces on infants’ attention-orienting may not yet be established. However, no studies have investigated the effect of faces on attention-orienting in infants older than 6 months of age.

The primary goal of the present study was to investigate the attention-orienting and attention-holding effects of faces in 4–8-month-old infants. We used a preferential looking-procedure with stimuli identical to those used by Libertus and Needham (2011). In this procedure, infants viewed eight pairs of stimuli consisting of a realistic, color photograph of a face presented adjacently to a brightly-colored toy (see Figure 1). Prior to the beginning of each trial, a moving green circle was presented in the center of the monitor in order to attract infants’ gaze to the center of the screen. When infants looked toward this center stimulus, a face–toy pair was presented. Each face–toy pair was presented for 10 seconds. There were eight trials total.

Figure 1.

Example of a face–toy pair displayed during test trials.

Attention-orienting was measured by calculating a proportion of first looks toward faces for each infant. We recorded the direction of infants’ first looks when a face–toy stimulus appeared. We then calculated the proportion of first looks across the eight trials for each infant. For example, if an infant looked toward the face first in 6 out of 8 trials, the proportion of first looks for that baby would be equal to .75. We reasoned that if faces influence infants’ attention-orienting response, the mean proportion of first looks across infants should be significantly greater than chance (.50).

Attention holding was measured by calculating a face preference score for each infant. The face-preference score was equal to an infants’ looking-time toward the faces divided by his or her total looking-time to the faces and toys across all eight trials. We reasoned that if faces held infants’ attention, the mean face preference score across infants would be significantly greater than chance (.50).

To investigate the robustness of infants’ face preference, we also examined whether the attention-holding effects of faces were stable throughout each trial. As noted by Cohen (1972), different patterns of looking behavior over the course of a trial can lead to the same overall looking-time result. For example, a pattern whereby an infant fixates on the face for 75% of a trial can be achieved by one long, stable fixation on the face stimulus and one short fixation on the object stimulus. Alternatively, that pattern could result from an infant making multiple shifts of attention between the stimuli with the majority of fixation time on the face stimulus. Thus, in the present study, we investigated whether faces held infants’ attention in a stable manner throughout the trial by examining face preference scores at 1-second intervals within each 10-second trial. If the attention-holding effects of faces were stable, we would expect to see face-preference scores greater than chance (.50) at each interval.

Finally, we explored whether sitting-ability would be related to any potential developmental differences in the effect of faces on these attentional processes. Libertus and Needham’s (2011) findings suggest there is a link between motor development and infants’ visual preferences for faces. Additionally, around 5–7 months, holistic face processing follows a U-shaped developmental curve (Cashon & Cohen, 2004), which has recently been associated with infants’ sitting abilities (Cashon, Ha, Allen, & Barna, 2012). Non-sitting infants (stage 1) and expert-sitting infants (stage 4) were found to process upright faces holistically, whereas infants who were learning to sit (stages 2 and 3) were found to regress to featural processing. Given the connection between physical experience and face perception found by Libertus and Needham (2011) and Cashon et al. (2012), we were interested in whether the development of sitting-ability would be associated with developmental changes in infants’ attention-orienting and attention-holding responses toward faces.

Method

Participants

Sixty-four full-term 4–8-month-old infants with normal hearing and vision participated. Participants included seven 4-month-olds (M = 17.88 weeks., SD = 1.31, 3 males, 4 females), 14 5-month-olds (M = 22.94 weeks, SD = 0.82, 10 males, 4 females), 19 6-month-olds (M = 26.36 weeks, SD = 1.30, 11 males, 8 females), 17 7-month-olds (M = 30.34 weeks, SD = 1.25, 4 males, 13 females), and seven 8-month-olds (M = 35.08 weeks, SD = 1.04, 3 males, 4 females). Data from three additional infants (two 4-month-olds and one 8-month-old) were excluded for not paying attention at all for four or more trials during the procedure. Participants were recruited using contact information obtained from local metropolitan birth records provided by the state. Families received a letter in the mail, along with a follow-up phone call inviting them to participate in the study. Infants received t-shirts or bibs for their participation.

Stimuli & apparatus

Stimuli consisted of eight pairs of colored pictures of faces and common infant toys displayed side-by-side on a white background (see Figure 1). The stimuli were identical to those used by Libertus and Needham (2011) and came from the NimStim face database (Tottenham et al., 2009). A Litemate III spot photometer was used to measure the luminance of the face and toy stimuli in candelas per meter squared (cd/m²). Similar luminance values were measured for faces (M= 38.88 cd/m²) and toys (M = 43.35 cd/m²). Face–toy pairs consisted of one of four neutral faces (2 male, 2 female) and one of four colorful infant toys. Each face and toy stimulus was shown twice, once on each side. Every trial contained a novel face–toy pair. Stimuli were presented sequentially on a 50-inch plasma display with a screen resolution of 1024 × 576 pixels. The faces and toys were comparable in size and subtended visual angles ranging from 13.13° by 13.03° to 13.13° by 15.26°. Faces and toys were separated by 2.86° to 4.76° of visual angle. The order of test stimuli was randomized across participants to control for any potential order effects.

Procedure

Sitting assessment

Prior to the face–toy preferential looking task, infants’ sitting-abilities were assessed using the procedure and classification system described in Cashon et al. (2012). Infants were placed in a sitting position by their parents on a blanket facing an experimenter, in profile to a camera. The experimenter timed how long infants were able to sit independently. Infants were classified into four stages: (1) non-sitters (n = 16, M_age = 21.46 weeks, SD_age = 4.08) were infants who were unable to sit upright for more than 2 seconds, (2) near-sitters (n = 17, M_age = 25.97, SD_age = 2.62) were infants who could sit upright or in a tripod position (that is, supporting themselves with their arms) for between 2 and 10 seconds, (3) new sitters (n = 17, M_age = 28.48, SD_age = 3.71) were infants who could sit upright independently for at least 10 seconds and have been doing so for less than 4 weeks, according to parents’ reports, and (4) expert sitters (n = 14, M_age = 31.40, SD_age = 3.40) were infants who could sit upright independently for at least 10 seconds and have been doing so for more than 4 weeks, according to parents’ reports.

Preferential-looking task

Following the sitting assessment, participants were seated on their parents’ laps approximately 120 cm away from the display in a dimly lit room. Parents were instructed not to interact with their infants during testing to avoid parental influence on their children’s looking-behaviors. An experimenter, who was seated in an adjacent room, observed infants on a 15” JVC closed-circuit monitor connected to a Canon VC-C50i camera, which was concealed below the center of the stimulus display. The experimenter used a Macintosh Power Mac G5, running Habit X1.0 software (Cohen, Atkinson, & Chaput, 2004), to present the stimuli. To begin the experiment, an attention-getting video consisting of a green, pulsating circle on a black background accompanied by a “ding” sound was presented to attract the participant’s attention to the center of the display. Once the infant looked toward the attention-getting video, the experimenter pressed the “enter” key on the keyboard to trigger the presentation of the first face–toy test trial. Infants were presented eight test trials, each with a fixed trial length of 10 s. Between test trials, the attention-getting video appeared briefly to attract the infant’s attention back to the center of the display. A video of each testing session was saved to DVD in order to code data offline. A naïve, independent experimenter coded left–right looks by observing the corneal reflections of infant participants frame-by-frame on each trial using Preferential Looking Coder (Version 1.3.2) coding software (Libertus, 2008). A second, naïve experimenter coded 20% of the videos for reliability purposes. The inter-rater reliability was r = .95.

Results

Attention-orienting

To measure the effect of faces on infants’ attention-orienting response, we calculated the proportion of first looks toward faces across eight trials. A one-sample t-test indicated that the overall mean proportion of infants’ first looks toward faces (M = .53, SD = .13) was not significantly different from chance (.50), t(63) = 1.67, p = .100, d = .23, 2-tailed.

To test whether there were any effects of sitting-ability, age or test trial on infants’ attention-orienting response toward faces, we analysed the mean proportion of first looks toward the face using a mixed 4 (sitting stage: 1–4) × 8 (trials: 1–8) analysis of covariance (ANCOVA) with age in weeks as a covariate.¹ No significant effects were found for age, F(1,52) = .02, p = .876, sitting stage, F(3,52) = .53, p = .662, or trials, F(1,52) = 1.39, p = .207. However, age and sitting stage were found to be significantly correlated, r(62) = .72, p < .001, thus a subsequent analysis was run without age as a covariate. Again, sitting-stage and trials were not significant.

Finally, to investigate whether infants demonstrated an orienting response to faces greater than chance on any trial. We analysed the mean proportion of first looks toward the face for each trial using one-sample t-tests. As illustrated in Figure 2, none of these means differed significantly from chance (.50) (all p’s > .25).

Figure 2.

Mean proportion of first looks toward faces (black bars) and toys (grey bars). Infants’ first looks were random across all trials, with no significant differences from chance (.50) on any trial, all p’s > .25.

Attention-holding

To determine whether faces were better than toys at holding infants’ attention, face preference scores ( $\frac{F a c e}{F a c e + T o y}$ ) were analysed. Unlike the null findings obtained for the attention orienting responses, the overall mean face preference score (M = .63, SD = .14) was found to be significantly greater than chance (.50), t(63) = 7.46, p < .001, d = .92, 2-tailed.

Mean face-preference scores for each sitting-stage across time are illustrated in Figure 3. To test whether there were any effects of sitting-ability, age, and time interval on mean face preference scores, we analysed the mean preference scores using a mixed 4 (sitting stage: 1–4) × 10 (time interval: 1–10) ANCOVA with age in weeks as a covariate. Mauchly’s test indicated that the assumption of sphericity had been violated (χ ² (44) = 778.06, p < .001); therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (∊ = .209). Age was not a significant covariate, F(1, 59) = .69, p = .410, and the analysis revealed no significant effects for time interval, F(1.88, 111.15) = 2.20, p = .118, or sitting stage, F(3, 59) = 1.10, p = .355. The removal of age as a covariate did not affect this pattern.

Figure 3.

Mean face preference scores by sitting stage, measured at 1-second intervals, and averaged across trials. Error bars represent standard errors.

After collapsing all the data across sitting-stage and age, we then compared mean face preference scores at each time interval to chance (.50). Mean face preference scores ranged from .60 to .63. One-sample t-tests revealed that at each time interval infants showed a statistically significant preference for faces (all p’s < .001, 2-tailed).

Discussion

The present study demonstrates that faces are effective in holding 4- to 8-month-old infants’ attention regardless of age and sitting-stage, but faces failed to initially capture infants’ attention at the beginning of trials. While the orienting data show that the direction of the first look is random, the attention-holding measure indicates that infants display a face-preference in the first time interval, and continue to do so throughout the remainder of each trial. These findings indicate that once the infant attends to the face stimulus, the face has a greater ability to hold infants’ attention relative to the objects.

The present findings extend previous findings addressing attention-holding effects of faces with infants around 5–6 months of age to infants up to 8 months of age. Together, our findings combined with Gliga et al. (2009), Di Giorgio et al. (2012), and Libertus and Needham (2011) suggest that infants’ face preferences are robust and well established in 4- to 8-month-olds. However, evidence for attention-orienting effects is inconsistent. Using the direction of infants’ first looks as our measure of orienting, we did not find any evidence of an orienting effect of faces. At first glance, the results of the current study appear to contrast those of Libertus and Needham, who reported an orienting effect of faces in 5-month-olds using identical stimuli. The current study differs from Libertus and Needham in how orienting was measured. Libertus and Needham defined orienting as a face preference within the 2nd second of looking time, but they did not report the direction of first looks. Our looking-time data during the 2nd second replicate the findings of Libertus and Needham. Additionally, two other studies that did use first looks as a measure of orienting produced inconsistent results. Gliga et al (2009) demonstrated an orienting effect in 6-month-olds when presented with a face within an array of full-color, common objects. Di Giorgio et al. (2012), in contrast, did not find an orienting effect in 3- or 6-month-olds, using a similar task, but with black and white stimuli. Thus, it appears that the face-orienting effect is less robust than the attention-holding effect in the middle of the first year. Further research needs to be conducted to shed light on the effect of faces on attention-orienting throughout development.

With respect to age and sitting-ability, no significant relations were found between either of these variables and the attention-orienting or attention-holding measures. Although prior research has shown that motor development relates to face perception and attention (Cashon et al., 2012; Libertus & Needham, 2011), a relation between infants’ learning to sit upright and their attention to faces was not found in the present study. It is possible that infants’ face preference is stable at this point and not susceptible to the influence of physical experience. Physical experience may only affect perception and attention during periods of developmental change (Cashon & DeNicola, 2011).

In the future, studies should explore the developmental effects of face inversion on infants’ attentional processes. Gliga et al. (2009) began to address the question of inversion in their study of 6-month-olds. They found that at this age, upright and inverted faces were equal in attracting infants’ first looks; however infants fixated more often to upright than inverted faces. This issue still needs to be addressed from a developmental perspective. Inversion has been shown to affect holistic processing of faces at around 7–8 months, but not in infants around 4 months of age (Cohen & Cashon, 2001; Ferguson, Kulkofsky, Cashon, & Casasola, 2009). Thus, it may be the case that the attention-orienting and attention-holding effects of upright and inverted faces would be comparable around 4 months of age, but by 7–8 months, upright and inverted faces would affect infants’ attention differently.

In conclusion, the effect of faces on the attention-holding process in infants is robust in 4- to 8-month-olds, but the effect of faces on the attention-orienting response is untenable. Faces fail to capture infants’ initial orienting response when measured by the direction of infants’ first looks. However, by the 1-second time intervals, infants display a face preference that is maintained for the remainder of the trial. Neither sitting ability nor age had any effects on this pattern. The findings suggest that faces have differential effects on these two attentional processes; While the findings do not show an attention orienting effect toward faces, they clearly show that faces have a powerful influence over holding infants’ attention in this age range.

Footnotes

Acknowledgements

The authors would like to thank the students in the University of Louisville Infant Cognition Laboratory for their help with data collection and coding. We thank the parents and infants who participated in this research and the Kentucky Cabinet for Health and Family Services for their assistance. We would also like to thank Klaus Libertus and Amy Needham for generously sharing their stimuli and coding systems with us.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported in part by University of Louisville Intramural Research Incentive Grant (#50627).

Note

References

Cashon

C. H.

Cohen

L. B.

(2004). Beyond U-shaped development in infants’ processing of faces. Journal of Cognition and Development, 5, 59–80. doi:10.1207/s15327647jcd0501_4

Cashon

C. H.

DeNicola

C. A.

(2011). Is perceptual narrowing too narrow? Journal of Cognition and Development, 12(2), 159–162. doi:10.1080/15248372.2011.563483

Cashon

C. H.

Allen

C. A.

Barna

A. C.

(2012). A U-shaped relation between sitting ability and upright face processing in infants. Child Development. Advance online publication. doi:10.1111/cdev.12024

Cohen

L. B.

(1972). Attention-getting and attention-holding processes of infant visual preferences. Child Development, 43, 869–879. doi:10.2307/1127638

Cohen

L. B.

(1976). Habituation of infant visual attention. In: Tighe

T. J.

Leaton

R. N.

(Eds.), Habituation: Perspectives from child development, animal behavior, and neurophysiology. (pp. 207–238). Hillsdale, NJ: L. Erlbaum Associates.

Cohen

L. B.

Atkinson

D. J.

Chaput

H. H.

(2004). HabitX: A new program for obtaining and organizing data in infant perception and cognition studies (Version 1.0). Austin: Texas.

Cohen

L. B.

Cashon

C. H.

(2001). Do 7-month-old infants process independent features or facial configurations? Infant and Child Development, 10, 83–92. doi:10.1002/icd.250

Di Giorgio

Turati

Altoè

Simion

(2012). Face detection in complex visual displays: An eye-tracking study with 3- and 6-month-old infants and adults. Journal of Experimental Child Psychology, 113, 66–77. doi:10.1016/j.jecp.2012.04.012

Easterbrook

M. A.

Kisilevsky

B. S.

Muir

D. W.

Laplante

D. P.

(1999). Newborns discriminate schematic faces from scrambled faces. Canadian Journal of Experimental Psychology/Revue canadienne de psychologie expérimentale, 53, 231–241. doi:10.1037/h0087312

10.

Fantz

R. L.

(1961). The origin of form perception. Scientific American, 204(5), 66–72. doi:10.1038/scientificamerican0561-66

11.

Fantz

R. L.

(1963). Pattern vision in newborn infants. Science, 140, 296–297. doi:10.1126/science.140.3564.296

12.

Ferguson

K. T.

Kulkofsky

Cashon

C. H.

Casasola

(2009). The development of specialized processing of own-race faces. Infancy, 14, 263–284. doi:10.1080/15250000902839369

13.

Frank

M. C.

Vul

Johnson

S. P.

(2009). Development of infants’ attention to faces during the first year. Cognition, 110, 160–170. doi:10.1016/j.cognition.2008.11.010

14.

Gliga

Elsabbagh

Andravizou

Johnson

(2009). Faces attract infants’ attention in complex displays. Infancy, 14, 550–562. doi:10.1080/15250000903144199

15.

Goren

C. C.

Sarty

P. Y. K.

(1975) Visual following and pattern discrimination of face-like stimuli by newborn infants. Pediatrics, 56, 544–549.

16.

HabitX: A new program for obtaining and organizing data in infant perception and cognition studies (Version 1.0) [Computer software]. Austin: University of Texas.

17.

Johnson

M. H.

Dziurawiec

Ellis

Morton

(1991). Newborns’ preferential tracking of face-like stimuli and its subsequent decline. Cognition, 40, 1–19. doi:10.1016/0010-0277(91)90045-6

18.

Libertus

. (2008). Preferential Looking Coder (Version 1.3.2) [Coding Software]. Durham, NC: Duke University. Retrieved October 9, 2009. Available from klaus.libertus@duke.edu

19.

Libertus

Needham

(2011). Reaching experience increases face preference in 3-month-old infants. Developmental Science, 14, 1355–1364. doi:10.1111/j.1467-7687.2011.01084.x

20.

Morton

Johnson

M. H.

(1991). CONSPEC and CONLERN: A two-process theory of infant face recognition. Psychological Review, 98, 164–181. doi:10.1037/0033-295X.98.2.164

21.

Preferential Looking Coder (Version 1.3.2) [Coding Software]. Durham, NC: Duke University. Available from klaus.libertus@duke.edu

22.

Tottenham

Tanaka

Leon

A. C.

McCarry

Nurse

Hare

T. A.,

… Nelson,

(2009). The NimStim set of facial expressions: judgments from untrained research participants. Psychiatry Research, 168, 242–249. doi:10.1016/j.psychres.2008.05.006

23.

Valenza

Simion

Cassia

Umiltà

(1996). Face preference at birth. Journal of Experimental Psychology: Human Perception and Performance, 22, 892–903. doi:10.1037/0096-1523.22.4.892