114 VOLUME 12, NUMBER 4, AUGUST 2003

Published by Blackwell Publishing Inc.

References

Fiser, J., & Aslin, R.N. (2002). Statistical learning ofnew visual feature combinations by infants.

Proceedings of the National Academy of Sciences,USA

,

99

, 15822–15826.Gomez, R.L., & Gerken, L. (1999). Artificial gram-

mar learning by 1-year-olds leads to specificand abstract knowledge.

Cognition

,

70

, 109–135.

Hauser, M., Newport, E.L., & Aslin, R.N. (2001).Segmentation of the speech stream in a non-human primate: Statistical learning in cotton-top tamarins.

Cognition

,

78

, B41–B52.Kirkham, N.Z., Slemmer, J.A., & Johnson, S.P.

(2002). Visual statistical learning in infancy:Evidence of a domain general learning mecha-nism.

Cognition

,

83

, B35–B42.Maye, J., Werker, J.F., & Gerken, L. (2002). Infant

sensitivity to distributional information can af-

fect phonetic discrimination.

Cognition

,

82

,B101–B111.

Morgan, J.L., Meier, R.P., & Newport, E.L. (1987).Structural packaging in the input to languagelearning: Contributions of intonational andmorphological marking of phrases to the ac-quisition of language.

Cognitive Psychology

,

19

,498–550.

Newport, E.L., & Aslin, R.N. (2000). Innately con-strained learning: Blending old and new ap-proaches to language acquisition. In S.C.Howell, S.A. Fish, & T. Keith-Lucas (Eds.),

Pro-ceedings of the 24th Boston University Conferenceon Language Development

(pp. 1–21). Somer-ville, MA: Cascadilla Press.

Pinker, S., & Bloom, P. (1990). Natural languageand natural selection.

Behavioral and Brain Sci-ences

,

13

, 707–784.Saffran, J.R. (2001a). The use of predictive depen-

dencies in language learning.

Journal of Mem-ory and Language

,

44

, 493–515.

Saffran, J.R. (2001b). Words in a sea of sounds: Theoutput of statistical learning.

Cognition

,

81

,149–169.

Saffran, J.R. (2002). Constraints on statistical lan-guage learning.

Journal of Memory and Lan-guage

,

47

, 172–196.

Saffran, J.R., Aslin, R.N., & Newport, E.L. (1996).Statistical learning by 8-month-old infants.

Sci-ence

,

274

, 1926–1928.

Saffran, J.R., Johnson, E.K., Aslin, R.N., & New-port, E.L. (1999). Statistical learning of tonesequences by human infants and adults.

Cogni-tion

,

70

, 27–52.

Saffran, J.R., & Thiessen, E.D. (2003). Pattern in-duction by infant language learners.

Develop-mental Psychology

,

39

, 484–494.

Saffran, J.R., & Wilson, D.P. (2003). From syllablesto syntax: Multi-level statistical learning by 12-month-old infants.

Infancy

,

4

, 273–284.

The Origins of Pictorial Competence

Judy S. DeLoache,

1

Sophia L. Pierroutsakos, and David H. Uttal

Department of Psychology, University of Virginia, Charlottesville, Virginia (J.S.D.); Department of Psychology, Furman University, Greenville, South Carolina (S.L.P.); and Department of Psychology, Northwestern University, Evanston, Illinois

(D.H.U.)

AbstractPictorial competence

, whichrefers to the many factors in-volved in perceiving, interpret-ing, understanding, and usingpictures, develops graduallyover the first few years of life.Although experience is not re-quired for accurate perceptionof pictures, it is necessary forunderstanding the nature ofpictures. Infants initially re-spond to depicted objects as ifthey were real objects, and tod-dlers are remarkably insensi-tive to picture orientation. Onlygradually do young childrenfigure out the nature of picturesand how they are used.

Keywords

symbolic development; pic-ture perception

As philosophers, new and old,have emphasized, humans are “the

symbolic species” (Deacon, 1997),and symbolization is the “mostcharacteristic mental trait of [hu-mans]” (Langer, 1942, p. 72). Just asthe emergence of the symbolic ca-pacity in the course of evolution ir-revocably transformed the humanspecies, so too does the develop-ment of symbolic functioningtransform young children. The ca-pacity for symbolization vastly ex-pands their intellectual horizons,liberating them from the con-straints of time and space and en-abling them to acquire informationabout reality without directly expe-riencing it.

All children growing up any-where in the world must master awide variety of symbol systemsand symbolic artifacts for full par-ticipation in their society. Our re-search has focused on how youngchildren begin to understand andexploit the informational potentialof various symbolic objects, includ-ing models, maps, and pictures.

We define a symbolic artifact assomething that someone intends tostand for something other than it-self (DeLoache, 1995). Thus, virtu-ally anything can serve as a sym-bol, and virtually any concept thatone has can be symbolized, but thesymbol is always different in someway from that which it represents.What makes something symbolic ishuman intention; an entity be-comes a symbol only as the resultof a person using it to denote or re-fer to something.

THE CHALLENGE OFDUAL REPRESENTATION

Although mastering symbols isa universal task, it is not an easyone. A formidable challenge toyoung children in developing com-petence with symbols stems fromthe inherently dual nature of sym-bols; every symbolic artifact is anobject in and of itself, and at thesame time it also stands for some-thing other than itself. To under-stand and use a symbol,

dual repre-sentation

is necessary—one mustmentally represent both facets ofthe symbol’s dual reality, both itsconcrete characteristics and its ab-stract relation to what it stands for

Copyright © 2003 American Psychological Society

CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE 115

(DeLoache, Pierroutsakos, & Tro-seth, 1996).

It is quite difficult for youngchildren to simultaneously hold inmind the two aspects of symbolicobjects. The younger the child, themore he or she tends to focus onthe concrete object itself, conse-quently failing to appreciate its ab-stract relation to what it stands for.To achieve dual representationwith respect to a given symbolicobject, children must avoid becom-ing “captured” by the object itselfand keep in mind both the objectand its referent. Research on thedevelopment of children’s under-standing of pictures reveals thatthese apparently simple symbolspresent challenges to the newestmembers of the symbolic species.

PICTURES AND THE INNOCENT OR

INTELLIGENT EYE

A long-standing debate in psy-chology and other fields has cen-tered on the degree to which expe-rience is necessary to perceive andunderstand pictures: Do infantsand young children look at pic-tures with an innocent eye (Gom-brich, 1969) or with an intelligenteye (Gregory, 1970)? In otherwords, do children have to learnhow to read pictures, how to inter-pret the code that relates picturesto the world they represent, as arthistorian Gombrich (1969) and phi-losopher Goodman (1976) argued?Are pictures cultural artifacts thatmembers of a society must learn tointerpret in context, as argued byanthropologists such as Dere-gowski (1989)? Or was Gibson(1979) correct that special skills arenot needed for dealing with pic-tures because they provide muchof the same information that isavailable from the real world?

As with many long-standing de-bates, neither the answer nor the

appropriate question is so simple.For one thing, the complex natureof even the simplest pictures is of-ten overlooked. Ittelson (1996)pointed out that specifying whatthe term “picture” properly refersto is surprisingly difficult. Giventhat pictures themselves are com-plex, it follows that achieving a fullunderstanding of them is quitecomplex. To foreshadow our con-clusion, we maintain that theyoung eye is never wholly inno-cent, but it is only as intelligent asthe head in which it resides.

PICTORIAL COMPETENCE

We use the term

pictorial compe-tence

to encompass the many fac-tors that are involved in perceiv-ing, interpreting, understanding,and using pictures, ranging fromthe straightforward perception andrecognition of simple pictures tothe most sophisticated understand-ing of the conventions and tech-niques of highly complex ones (De-Loache et al., 1996). Full pictorialcompetence involves both perceptualabilities and conceptual knowledge.In perceiving and interpreting a pic-ture, a viewer not only sees the rep-resentation—the picture surface—but also “sees through” it to its ref-erent. At the same time, the viewermust understand and keep in mindthe nature of the relation betweenrepresentation and referent. Fi-nally, pictorial competence also in-cludes pragmatic knowledge abouthow pictures are produced andused.

There is ample evidence thatyoung infants, even newborns,possess some elements of pictorialcompetence. (See DeLoache et al.,1996, for a summary.) For onething, they can discriminate be-tween actual objects and depictionsof objects. For another, infants canrecognize familiar objects and peo-ple in photographs or drawings.

Clearly, pictorial experience is notnecessary for accurate picture per-ception. But this precocity in pic-ture perception is only the begin-ning of the story.

GRASPING THE NATUREOF PICTURES

Despite their ability to perceivepictures and discriminate depictedfrom real objects, young infants donot understand the nature of pic-tures nor the differences betweenpictures and their referents. Thisconclusion comes from an ongoingseries of studies in which we areobserving infants as they are al-lowed to explore pictures of famil-iar types of objects (DeLoache,Pierroutsakos, Uttal, Rosengren, &Gottlieb, 1998; Pierroutsakos & De-Loache, 2003). Each infant sits in ahigh chair, and a book is placed onthe tray directly in front of him orher. The specially constructed bookswe use contain highly realisticcolor photographs of single objects.

A person who can perceive pic-tures and also understands theirnature would simply look at thedepicted objects. That is not, how-ever, what the 9-month-old infantsin our research do. Instead, almostevery infant we have tested in sev-eral studies has at least once manu-ally explored the depicted objects.As illustrated in Figure 1, babiestouch, rub, and strike at the surfaceof the pictures, and they frequentlymake grasping motions as if tryingto pick up the depicted objects.

Some babies are notably persis-tent, making repeated attempts topluck the objects from the page;others react to some but not all ofthe pictures; and a few are satisfiedwith a lone attempt. The impres-sion we have watching the infantsin our research is that some ofthem really think they should beable to grasp and pick up the de-picted objects. Others do not seem

116 VOLUME 12, NUMBER 4, AUGUST 2003

Published by Blackwell Publishing Inc.

to be so sure, but nevertheless giveit a try.

It is important to note that thisbehavior does not reflect a failureto discriminate between picturesand objects. When a real object ispresented side by side with a pho-tograph of that object, infants al-most always reach for the objectfirst, indicating clear discriminationbetween two- and three-dimensionalstimuli. At the same time, the morea picture resembles an actual ob-ject, the more infants try to interactwith it. As shown in Figure 2,

in-fants’ manual investigation of de-picted objects varies as a functionof the degree of pictorial realism(Pierroutsakos & DeLoache, 2003).Color photographs elicit substan-tially more manual investigationthan do less realistic pictures.

Infants’ manual exploration oftwo-dimensional images is not lim-ited to still pictures. In one study(Pierroutsakos & Troseth, 2003),9-month-olds seated close to a mon-itor viewed a video presentation inwhich a hand deposited an objecton a tabletop. The infants graspedat and tried to pluck the object offthe video screen, just as children inthe earlier studies did with pic-tures in books. When shown amoving object, such as a windupsnail slowly lumbering across thescreen, the children were particu-larly persistent at reaching for it.

The manual exploration of pic-tures that is so common and easilyelicited in 9-month-olds becomesincreasingly less common as theygrow older. In one of our studies(DeLoache et al., 1998), we com-pared how young children of dif-ferent ages interacted with pictures(see Fig. 3).

The level of manualexploration was highest among

9-month-olds and substantiallylower among 15-month-olds; virtu-ally none of the 19-month-olds in-vestigated the pictures manually.The decline in manual explorationamong the older children was ac-companied by an increase in a dif-ferent manual behavior—pointingat pictures, often accompanied byverbalizations. Thus, as children

Fig. 1. Infants’ manual investigation of pictures. The babies on the left and in the middle are manually investigating a color photo-graph of an object. In an excess of zeal, the third child has bent over to put his mouth on the nipple of a depicted baby bottle. Allthree of these children are to some extent treating the depicted objects as if they were real objects.

Fig. 2. The relation between infants’ manual exploration of pictures and the pictures’degree of realism (Pierroutsakos & DeLoache, 2003). Across the range from black-and-white (B & W) line drawings through color photographs, the more a depictedobject resembles a real object, the more infants manually explore the image by feel-ing, patting, and grasping at it.

Copyright © 2003 American Psychological Society

CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE 117

become older, they increasingly at-tempt to get another person toshare their interest in a depictedobject, instead of trying to interactdirectly with it. By the time chil-dren are 19 months of age, de-picted objects have become objectsof contemplation and communica-tion, not action.

Why do infants grasp at pic-tures? We think it is because theydo not understand the

nature

ofpictures. As we have noted, there isno question that they can accu-rately perceive a difference be-tween pictures and objects. Whatour research indicates is that in-fants do not yet understand the sig-nificance of that difference; they donot appreciate that pictures shareonly some of the attributes of theirreferents.

We thus think the manual explo-ration of pictures that we have doc-umented reflects a conceptual, nota perceptual, deficit. Infants do notknow

what kind of thing

a picture is.Not understanding the significanceof two-dimensionality, they re-spond to realistic pictures as if theywere three-dimensional objects. By19 months of age, the middle-classAmerican children we have stud-ied have come to understand the

crucial nature of the difference be-tween real objects and depicted ob-jects. (This developmental changewould presumably occur moreslowly in societies in which childrenreceive substantially less exposureto pictures.) Despite their perceptualand cognitive prowess, young chil-dren still have a substantial amountof pragmatic knowledge to acquireabout how pictures are used.

PICTURE ORIENTATION

Nothing seems more natural toadults than that a picture has a cor-rect orientation, and adults have astrong preference for viewing pic-tures right side up—with the de-picted entity oriented to the vieweras it would typically be in the world.Young children do not share thispreference.

Our research on young chil-dren’s sensitivity to picture orien-tation started with casual observa-tions and parental anecdotes oftoddlers happily looking at theirpicture books upside down—arather surprising observation,given that much younger infants

can discriminate orientation andprefer looking at some stimuli (fa-miliar faces) in their normal up-right orientation. Studies in our labhave verified that very young chil-dren have at best a weak prefer-ence for viewing pictures right sideup (DeLoache, Uttal, & Pierroutsa-kos, 2000; Pierroutsakos & De-Loache, 2003). If 18-month-olds arehanded an upside-down book tolook at, they usually study it in itsinverted orientation, not botheringto turn it around. If an adult holdsa picture book upside down whilereading to them, they typically donot protest or try to reorient it.Their indifference to orientation isspecific to pictures: If handed anupside-down object, they almostalways reorient it.

Young children’s indifference topicture orientation extends be-yond a lack of preference: In twostudies using different kinds of pic-tures, 18-month-olds were equallyaccurate at identifying pictures intheir upright and inverted orienta-tions. Furthermore, the time it tookthem to point to the correct itemdid not differ for upright and in-verted pictures (Pierroutsakos &DeLoache, 2003).

The 18-month-olds in this re-search had clearly not yet acquiredthe cultural convention of pictureorientation. Although they knewthat picture-book interactions in-volve pointing to and talking aboutpictures, they had not yet devel-oped a preference for looking atpictures in a particular orientation.An interesting question is to whatextent that development may arisefrom a shift from a piecemeal to amore integrated perception of pic-tures.

DEVELOPING AN INTELLIGENT EYE

The research summarized hereindicates that infants and very

Fig. 3. Mean number of times 9-month-olds, 15-month-olds, and 19-month olds in-teracted with pictures by manually exploring them versus pointing at them. Manualexploration of depicted objects decreased with age, presumably because as infantsgrow older, they figure out how depicted objects differ from real ones. At the sametime, they increasingly point to pictures, often while naming the depicted object orrequesting that someone else name it. From DeLoache, Pierroutsakos, Uttal, Rosen-gren, and Gottlieb (1998).

118 VOLUME 12, NUMBER 4, AUGUST 2003

Published by Blackwell Publishing Inc.

young children only gradually de-velop an intelligent eye for pic-tures. Very early on, infants possesssome aspects of pictorial compe-tence. However, it takes severalmonths of pictorial experience to ap-preciate the fact that a picture sharesonly a few of the qualities of its ref-erent. An appreciation for the prag-matics of picture use comes later,with toddlers gradually adopting theconvention that pictures are to belooked at in a particular orienta-tion. Full pictorial competence takesseveral more years to develop, asyoung children come to under-stand increasingly more about pic-tures (see DeLoache et al., 1996).

Our research leads us to bothagree and disagree with Lopes(1996) that “no picture is seen withan innocent eye, because we cometo pictures primed with beliefs, ex-pectations, and attitudes about sys-tems of representation” (p. 33). Weagree that no picture is ever seenwith an innocent eye, because evenyoung infants bring to bear what-ever knowledge they have aboutthe real-world entity depicted in apicture; they see through the pic-ture to its referent. However, thestatement lacks any developmentalperspective. Infants do not initiallyhave any conceptual knowledgeabout pictures, and the develop-ment of full pictorial competence—the emergence of the truly intelli-gent eye in the thinking brain—proceeds for several years.

CONCLUSION

The research summarized hereraises more questions that it an-swers—as should always be thecase. For example, how might in-fants’ manual exploration of pic-tures be influenced by perceptualfeatures of the objects depicted andby children’s past experience with

those objects? In the developmentof sensitivity to picture orientation,what role is played by changes inbasic perceptual processing, specif-ically, a general shift from rela-tively piecemeal processing tomore holistic processing of picto-rial stimuli?

The central issue underlying theresearch described here is the inter-play of perceptual experience andconceptual knowledge, a generalproblem of continuing interestthroughout psychology. An impor-tant topic in this domain—one ap-plicable to any age group—concernswhat kinds of pictorial displays fa-cilitate learning. We are currentlyexamining this general problem byinvestigating what and how infantsand toddlers learn from picturebooks—an issue that has neverbeen studied. Our initial studiesconcern whether new words andideas are learned and generalizedbetter from books containinghighly realistic pictures or frompictures less tied to specific refer-ents. Given our prior work, thereare reasons to expect either possi-bility might be correct. We expectthe result of this work to be of theo-retical and practical interest.

As a final note, we wish to makethe methodological point that re-searchers should never take it forgranted that pictorial stimuli cantell them about real-world behav-ior with real objects. “Generaliza-tions to and from studies using pic-tures as stimuli must be carefullyexamined” (Ittelson, 1996, p. 181).Many controversies, includingsome in developmental psychology,might be at least partly traceable todifferences in the effects of usingpictures and real objects as stimuli.

Recommended Reading

DeLoache, J.S. (2002). The symbol-mindedness of young children. InW. Hartup & R. Weinberg (Eds.),

The Minnesota Symposium on Child

Psychology: Vol. 32. In celebration ofthe 75th anniversary of the Instituteof Child Development

(pp. 73–101).Mahwah, NJ: Erlbaum.

DeLoache, J.S., Pierroutsakos, S.L.,Uttal, D.H., Rosengren, K.S., &Gottlieb, A. (1998). (See Refer-ences)

Ittelson, W.H. (1996). (See Refer-ences)

Note

1. Address correspondence to JudyS. DeLoache, Psychology Department,University of Virginia, P.O. Box 400400,Charlottesville, VA 22904; e-mail:jdeloache@virginia.edu.

References

Deacon, T.W. (1997).

The symbolic species: The co-evolution of language and the brain

. New York:Norton.

DeLoache, J.S. (1995). Early symbolic understand-ing and use. In D. Medin (Ed.),

The psychologyof learning and motivation, Vol. 33

(pp. 65–114).New York: Academic Press.

DeLoache, J.S., Pierroutsakos, S.L., & Troseth, G.L.(1996). The three ‘R’s of pictoral competence.In R. Vasta (Ed.),

Annals of child development,Vol. 12

(pp. 1–48). Bristol, PA: Jessica Kingsley.DeLoache, J.S., Pierroutsakos, S.L., Uttal, D.H.,

Rosengren, K.S., & Gottlieb, A. (1998). Grasp-ing the nature of pictures.

Psychological Science

,

9

, 205–210.DeLoache, J.S., Uttal, D.H., & Pierroutsakos, S.L.

(2000). What’s up? The development of an ori-entation preference for picture books.

Journalof Cognition and Development

,

1,

81–95.Deregowski, J. (1989). Real space and represented

space: Cross-cultural perspectives.

Behavioraland Brain Sciences

,

12

, 51–119.Gibson, J.J. (1979).

The ecological approach to visualperception

. Boston: Houghton Mifflin.Gombrich, E.H. (1969).

Art and illusion: A study inthe psychology of pictorial representation

. Prince-ton, NJ: Princeton University Press.

Goodman, N. (1976).

Languages of art: An approachto a theory of symbols

(2nd ed.). Indianapolis,IN: Hackett Publishing.

Gregory, R.L. (1970).

The intelligent eye

. New York:McGraw-Hill.

Ittelson, W.H. (1996). Visual perception of mark-ings.

Psychonomic Bulletin & Review

,

3

, 171–187.Langer, S.K. (1942).

Philosophy in a new key

. Cam-bridge, MA: Harvard University Press.

Lopes, D. (1996). Understanding pictures. Oxford,England: Clarendon Press.

Pierroutsakos, S.L., & DeLoache, J.S. (2003). In-fants’ manual investigation of pictured objectsvarying in realism. Infancy, 4, 141–156.

Pierroutsakos, S.L., & Troseth, G.L. (2003). Videoverite: Infants’ manual investigation of objectsdepicted on video. Infant Behavior and Develop-ment, 26, 183–189.