The emergence of object‐based visual attention in infancy: A role for family socioeconomic status and competing visual features

The development of spatial visual attention has been extensively studied in infants, but far less is known about the emergence of object‐based visual attention. We tested 3–5‐ and 9–12‐month‐old infants on a task that allowed us to measure infants’ attention orienting bias toward whole objects when they competed with color, motion, and orientation feature information. Infants’ attention orienting to whole objects was affected by the dimension of the competing visual feature. Whether attention was biased toward the whole object or its salient competing feature (e.g., “ball” or “red”) changed with age for the color feature, with infants biased toward whole objects with age. Moreover, family socioeconomic status predicted feature‐based attention in the youngest infants and object‐based attention in the older infants when color feature information competed with whole‐object information.     

Infants' Individuation of Rigid and Plastic Objects Based on Shape

Recent research suggests that 12‐month‐old infants use shape to individuate the number of objects present in a scene. This study addressed the question of whether infants would also rely on shape when shape is only a temporary attribute of an object. Specifically, we investigated whether infants realize that shape changes reliably indicate identity changes only in the case of rigid objects, but not in the case of deformable plastic objects. Twelve‐month‐old infants observed how either a rigid or a plastic object was placed in a box. When searching the box, they retrieved either an object with the same (no‐switch event) or with a different shape (switch event). Infants correctly inferred two distinct objects in the switch event in the case of rigid objects, but not in the case of plastic objects. A control experiment confirmed that this result was not due to a lack of salience of the shape transformation. Thus, infants' re‐searching behavior indicated that they viewed shape as being diagnostic in the individuation process of rigid objects only.     

Rules Infants Look By: Testing the Assumption of Transitivity in Visual Salience

What drives infants’ attention in complex visual scenes? Early models of infant attention suggested that the degree to which different visual features were detectable determines their attentional priority. Here, we tested this by asking whether two targets—defined by different features, but each equally salient when evaluated independently—would drive attention equally when pitted head‐to‐head. In Experiment 1, we presented 6‐month‐old infants with an array of Gabor patches in which a target region varied either in color or spatial frequency from the background. Using a forced‐choice preferential‐looking method, we measured how readily infants fixated the target as its featural difference from the background was parametrically increased. Then, in Experiment 2, we used these psychometric preference functions to choose values for color and spatial frequency targets that were equally salient (preferred), and pitted them against each other within the same display. We reasoned that if salience is transitive, then the stimuli should be iso‐salient and infants should therefore show no systematic preference for either stimulus. On the contrary, we found that infants consistently preferred the color‐defined stimulus. This suggests that computing visual salience in more complex scenes needs to include factors above and beyond local salience values.     

To Grasp or Not to Grasp: Infants’ Actions Toward Objects and Pictures

We conducted two experiments to address questions over whether 9‐month‐old infants believe that objects depicted in realistic photographs can be picked up. In Experiment 1, we presented 9‐month‐old infants with realistic color photographs of objects, colored outlines of objects, abstract colored “blobs,” and blank pages. Infants most commonly rubbed or patted depictions of all types. They also showed significantly more grasps toward the realistic photographs than toward the colored outlines, blobs, and blank pages, but only 24% of infants directed grasping exclusively at the photographs. In Experiment 2, we further explored infants’ actions toward objects and pictures while controlling for tactile information. We presented 9‐month‐old infants with objects and pictures of objects under a glass cover in a false‐bottom table. Although there were no significant differences between the proportion of rubs and pats infants directed toward the objects versus the photographs, infants exhibited significantly more grasping toward the objects than the photographs. Together, these findings show that 9‐month‐old infants largely direct appropriate actions toward realistic photographs and real objects, indicating that they perceive different affordances for pictures and objects.     

Word Level Stress and Lexical Processing in 17‐Month‐Old Infants

Detailed representations enable infants to distinguish words from one another and more easily recognize new words. We examined whether 17‐month‐old infants encode word stress in their familiar word representations. In Experiment 1, infants were presented with pairs of familiar objects while hearing a target label either properly pronounced with the correct stress (e.g., baby /’be?bi/) or mis‐pronounced with the incorrect stress pattern (e.g., baby /be?’bi/). Infants mapped both the correctly stressed and mis‐stressed labels to the target objects; however, they were slower to fixate the target when hearing the mis‐stressed label. In Experiment 2, we examined whether infants appreciate that stress has a nonproductive role in English (i.e., altering the stress of a word does not typically signal a change in word meaning) by presenting infants with a familiar object paired with a novel object while hearing either correctly stressed or mis‐stressed familiar words (Experiment 2). Here, infants mapped the correctly stressed label to the familiar object but did not map the mis‐stressed label reliably to either the target or distractor objects. These findings suggest that word stress impacts the processing of familiar words, and infants have burgeoning knowledge that altering the stress pattern of a familiar word does not reliably signal a new referent.     

Learning While Babbling: Prelinguistic Object‐Directed Vocalizations Indicate a Readiness to Learn

Two studies illustrate the functional significance of a new category of prelinguistic vocalizing—object‐directed vocalizations (ODVs)—and show that these sounds are connected to learning about words and objects. Experiment 1 tested 12‐month‐old infants’ perceptual learning of objects that elicited ODVs. Fourteen infants’ vocalizations were recorded as they explored novel objects. Infants learned visual features of objects that elicited the most ODVs but not of objects that elicited the fewest vocalizations. Experiment 2 assessed the role of ODVs in learning word–object associations. Forty infants aged 11.5 months played with a novel object and received a label either contingently on an ODV or on a look alone. Only infants who received labels in response to an ODV learned the association. Taken together, the findings suggest that infants’ ODVs signal a state of attention that facilitates learning.     

Follow My Gaze: Face Race and Sex Influence Gaze‐Cued Attention in Infancy

Using the eye gaze of others to direct one's own attention develops during the first year of life and is thought to be an important skill for learning and social communication. However, it is currently unclear whether infants differentially attend to and encode objects cued by the eye gaze of individuals within familiar groups (e.g., own race, more familiar sex) relative to unfamiliar groups (e.g., other race, less familiar sex). During gaze cueing, but prior to the presentation of objects, 10‐month‐olds looked longer to the eyes of own‐race faces relative to 5‐month‐olds and relative to the eyes of other‐race faces. After gaze cueing, two objects were presented alongside the face and at both ages, infants looked longer to the uncued objects for faces from the more familiar‐sex and longer to cued objects for the less familiar‐sex faces. Finally, during the test phase, both 5‐ and 10‐month‐old infants looked longer to uncued objects relative to cued objects but only when the objects were cued by an own‐race and familiar‐sex individual. Results demonstrate that infants use face eye gaze differently when the cue comes from someone within a highly experienced group.     

Anticipatory Eye Movements Reveal Infants' Auditory and Visual Categories

We introduce a new paradigm for the assessment of auditory and visual categories in 6‐month‐old infants using a 2‐alternative anticipatory eye‐movement response. Infants were trained by 2 different methods to anticipate the location of a visual reinforcer at 1 of 2 spatial locations (right or left) based on the identity of 2 cuing stimuli. After a training phase, infants were presented with a series of generalization trials in which novel (untrained) stimuli served as the cue to the anticipatory eye movement. Four experiments illustrated that infants can learn the 2‐choice discriminative response during training. Infants also showed anticipatory eye movements to novel stimuli, indicating sensitivity to variations along a variety of stimulus dimensions (e.g., color, shape, orientation, spatial frequency, pitch, and duration). In addition, the paradigm can be used to assess categorization in individual infants, thereby revealing the stimulus dimensions to which infants naturally attend.     

Attention to Multiple Cues During Spontaneous Object Labeling

It is well established that 2‐year‐olds attribute a novel label to an object’s global shape rather than local features (i.e., parts). Although recent studies have found that younger infants also attend to global rather than local features when given a label, the test stimuli in these experiments confounded parts and shape by varying both or neither. With infants (16‐ and 24‐month‐olds) and adults, this experiment disentangled shape and parts with appropriate test objects. Results showed a clear development of a strategy incorporating multiple cues. Across three age groups, there was an increase in generalizing labels to objects matching the exemplar’s local and global features (parts, base, and shape), and a decrease to objects matching in only one local feature. We discuss these results in terms of a learned flexibility in using multiple cues to predict lexical categories.     

Infants' Manual Exploration of Pictorial Objects Varying in Realism

In previous research, we established that 9‐month‐old infants manually investigate pictured objects by hitting, rubbing, and grasping as if to pluck them off the page. This behavior suggests that infants do not understand the 2‐dimensional nature of pictures. Although they can perceive depth cues and distinguish pictures from objects, they do not appreciate the significance of these cues; that is, they do not realize how depicted objects differ from real ones. We report 2 studies that support the idea that infants' manual response to pictures is driven by the resemblance of depicted objects to the real objects they represent. In Study 1, we report that infants' manual investigation of pictures is directly related to how realistic they are: The more depicted objects look like real objects, the more manual investigation they evoke. In Study 2, we show that 9‐month‐old infants' manual behaviors are concentrated on depicted objects even when there are areas of greater perceptual contrast on the page. The results are discussed with respect to the early development of pictorial competence.     

Shake,Rattle, and … One or Two Objects? Young Infants' Use of Auditory Information to Individuate Objects

Most research on object individuation in infants has focused on the visual domain. Yet the problem of object individuation is not unique to the visual system, but shared by other sensory modalities. This research examined 4.5‐month‐old infants' capacity to use auditory information to individuate objects. Infants were presented with events in which they heard 2 distinct sounds, separated by a temporal gap, emanate from behind a wide screen; the screen was then lowered to reveal 1 or 2 objects. Longer looking to the 1‐ than 2‐object display was taken as evidence that the infants (a) interpreted the auditory event as involving 2 objects and (b) found the presence of only 1 object when the screen was lowered unexpected. The results indicated that the infants used sounds produced by rattles, but not sounds produced by an electronic keyboard, as the basis for object individuation (Experiments 1 and 2). Data collected with adult participants revealed that adults are also more sensitive to rattle sounds than electronic tones. A final experiment assessed conditions under which young infants attend to rattle sounds (Experiment 3). Collectively, the outcomes of these experiments suggest that infants and adults are more likely to use some sounds than others as the basis for individuating objects. We propose that these results reflect a processing bias to attend to sounds that reveal something about the physical properties of an object—sounds that are obviously linked to object structure—when determining object identity.     

Second‐Order Correlation Learning of Dynamic Stimuli: Evidence from Infants and Computational Modeling

We present two habituation experiments that examined 20‐ and 26‐month‐olds’ ability to engage in second‐order correlation learning for static and dynamic features, whereby learned associations between two pairs of features (e.g., P and Q, P and R) are generalized to the features that were not presented together (e.g., Q and R). We also present results from an associative learning mechanism that was implemented as an autoencoder parallel distributed processing (PDP) network in which second‐order correlation learning is shown to be an emergent property of the dynamics of the network. The experiments and simulation demonstrate that 20‐ and 26‐month‐olds as well as neural networks are capable of second‐order correlation learning in a category context for internal features of dynamic objects. However, the model predicts—and Experiment 3 demonstrates—that 20‐ and 26‐month‐olds are unable to encode second‐order correlations in a noncategory context for dynamic objects with internal features. It is proposed that the ability to learn second‐order correlations represents a powerful but as yet unexplored process for generalization in the first years of life.     

Fourteen‐Month‐Olds Know What “We” Have Shared in a Special Way

People often express excitement to each other when encountering an object that they have shared together previously in some special way. This study investigated whether 14‐month‐old infants know precisely what they have and have not shared in a special way (and with whom). In the experimental condition an adult and infant shared an object (the target) excitedly because it unexpectedly reappeared in several places. They then shared 2 other objects (the distractors) in a more normal fashion. Later, the adult reacted excitedly to a tray containing all 3 objects and then made an ambiguous request for the infant to hand “it” to her. There were 2 control conditions. In 1 of them, a different adult, who knew none of the 3 objects, made the ambiguous request. In the other control condition, the adult who made the request had previously experienced the objects only alone, while the infant looked on unengaged. Infants in the experimental condition chose the target object more often than the distractors and more often than they chose it in either control condition. These results demonstrate that 14‐month‐old infants can identify which one of a set of objects “we”—and not just I or you alone—have had a special experience with in the past.     

The Effects of Graded Occlusion on Manual Search and Visual Attention in 5‐ to 8‐Month‐Old Infants

Young infants may be limited in searching for hidden objects because they lack the means‐end motor skill to lift occluders from objects. This account was investigated by presenting 5‐ to 8‐month‐old infants with objects hidden behind transparent, semitransparent, and opaque curtains. If a means—end deficit explains search limitations, then infants should search no more for an object behind a transparent curtain than for objects behind semitransparent or opaque curtains. However, level of occlusion had a significant effect on manual search and visual attention. Infants retrieved and contacted the object more, contacted the curtain more, and looked away less with the transparent curtain than with the semi transparent or opaque curtains. Adding a time delay before allowing search and presenting a distraction after occlusion further depressed infants' behavior. The findings fail to support the means—end deficit hypothesis, but are consistent with the account that young infants lack object permanence.     

The Representation of Geometric Cues in Infancy

There is evidence that, from an early age, humans are sensitive to spatial information such as simple landmarks and the size of objects. This study concerns the ability to represent a particular kind of spatial information, namely, the geometry of an enclosed layout—an ability present in older children, adults, and nonhuman animals (e.g., Cheng, 1986; Hermer & Spelke, 1996). Using a looking‐time procedure, 4.5‐ to 6.5‐month‐olds were tested on whether they could distinguish among the corners of an isosceles triangle. On each trial, the target corner was marked by a red dot. The stimulus (triangle with dot) appeared from different orientations across trials, ensuring that only cues related to the triangle itself could be used to differentiate the corners. When orientations were highly variable, infants discriminated the unique corner (i.e., the corner with the smaller angle and two equal‐length sides) from a nonunique corner; they could not discriminate between the two nonunique corners. With less variable orientations, however, infants did discriminate between the nonunique corners of the isosceles triangle. Implications for how infants represent geometric cues are discussed.     

Investigating the Relation Between Infants' Manual Activity With Objects and Their Perception of Dynamic Events

We examined the relation between 6‐ and 7‐month‐old infants' (N = 60) manual activity with objects during free play and their perception of the features of dynamic, multimodal events. Infants were habituated to a single event in which a hand reached for and manipulated a colorful, multifeatured object, and a sound was heard (e.g., a hand squeezed a purple round object, causing a whistling sound) and then their response to events that involved a change in the appearance of the object, the action, or the sound was assessed. Infants responded least to changes in the appearance of the objects, and their sensitivity to this feature was related to their manual activity with objects during free play. Infants' responding to changes in the sound or action was unrelated to motor activity, suggesting that at this age motor achievements related to object exploration are associated with infants' perception of some, but not all, object features.     

Visual short‐term memory for overtly attended objects during infancy

We investigated limitations in young infants’ visual short‐term memory (VSTM). We used a one‐shot change detection task to ask whether 4‐ and 8.5‐month‐old infants (N = 59) automatically encode fixated items in VSTM. Our task included trials that consisted of the following sequence: first a brief (500 ms) presentation with a sample array of two items, next a brief (300 ms) delay period with a blank screen, and finally a test array (2,000 ms) identical to the sample array except that the color of one of the two items is changed. In Experiment 1, we induced infants to fixate one item by rotating it during the sample (the other item remained stationary). In Experiment 2, none of the items rotated. In both experiments, 4‐month‐old infants looked equally at the fixated item when it did and did not change color, providing no evidence that they encoded in VSTM the fixated item. In contrast, 8.5‐month‐old infants in Experiment 1 preferred the fixated item when it changed color from sample to test. Thus, 4‐month‐old infants do not appear to automatically encode fixated items in VSTM.     

Mothers' Verbal and Nonverbal Strategies in Relation to Infants' Object‐Directed Actions in Real Time and Across the First Three Years in Ethnically Diverse Families

We examined maternal behavioral strategies in relation to infants' object‐directed actions in real time and over developmental time in 206 mother–infant dyads from African American, Dominican immigrant, and Mexican immigrant backgrounds. Mothers were asked to share a set of beads and strings with their infants when children were 14, 24, and 36 months. We coded three types of maternal strategies—eliciting attention, instructive assistance, and encouragement—which could be expressed verbally (e.g., “look”, “turn it”, “good job!”) or physically (i.e., through gestures, hands‐on guidance, or transfer of objects). We also coded infants' unassisted bead‐stringing. Across ethnic groups and ages, mothers' hands‐on guidance and object transfer increased the likelihood that infants would follow with unassisted bead‐stringing during real‐time interaction. Over developmental time, mothers modified their strategies: They displayed fewer attention‐getting strategies and more encouragement across infant ages, and peaked in their provision of instructive assistance when infants were 24 months. Additionally, Mexican mothers displayed more nonverbal strategies (e.g., gestures, hands‐on guidance) than did African American and/or Dominican mothers, who displayed more verbal strategies (e.g., attention‐getting and encouraging language). Developmental and real‐time patterns in mother–infant object‐related interactions generalize across ethnicities, although mothers' emphases on specific strategies are culture specific.     

Hierarchical Processing in Seven‐Month‐Old Infants

Hierarchical structures are crucial to many aspects of cognitive processing and especially for language. However, there still is little experimental support for the ability of infants to learn such structures. Here, we show that, with structures simple enough to be processed by various animals, seven‐month‐old infants seem to learn hierarchical relations. Infants were presented with an artificial language composed of “sentences” made of three‐syllable “words.” The syllables within words conformed to repetition patterns based on syllable tokens involving either adjacent repetitions (e.g., dubaba) or nonadjacent repetitions (e.g., du ba du ) . Importantly, the sequence of word structures in each sentence conformed to repetition patterns based on word types (e.g., aba‐ abb ‐ abb ). Infants learned this repetition pattern of repetition patterns and thus likely a hierarchical pattern based on repetitions, but only when the repeated word structure was based on adjacent repetitions. While our results leave open the question of which exact sentence‐level pattern infants learned, they suggest that infants embedded the word‐level patterns into a higher‐level pattern and thus seemed to acquire a hierarchically embedded pattern.     

An object's categorizability impacts whether infants encode surface features into their object representations

Infants encode the surface features of simple, unfamiliar objects (e.g., red triangle) and the categorical identities of familiar, categorizable objects (e.g., car) into their representations of these objects. We asked whether 16–18-month-olds ignore non-diagnostic surface features (e.g., color) in favor of encoding an object's categorical identity (e.g., car) when objects are from familiar categories. In Experiment 1 (n = 18), we hid a categorizable object inside an opaque box. In No Switch trials, infants retrieved the object that was hidden. In Switch trials, infants retrieved a different object: an object from a different category (Between-Category-Switch trials) or a different object from the same category (Within-Category-Switch trials). We measured infants' subsequent searching in the box. Infants' pattern of searching suggested that only infants who completed a Within-Category-Switch trial as their first Switch trial encoded objects' surface features, and an exploratory analysis suggested that infants who completed a Between-Category-Switch trial as their first Switch trial only encoded objects' categories. In Experiment 2 (n = 18), we confirmed that these results were due to objects' categorizability. These results suggest infants may tailor the way they encode categorizable objects depending on which object dimensions are perceived to be task relevant.