Visual short-term memory for complex objects in 6- and 8-month-old infants

The impact of caregiver inhibitory control on infant visual working memory

Visual working memory (VWM) emerges in the first year of life and has far-reaching implications for academic and later life outcomes. Given that caregivers play a significant role in shaping cognitive function in children, it is important to understand how they might impact VWM development as early as infancy. The current study investigated whether caregivers' efficiency of regulating inhibitory control was associated with VWM function in their infants. Eighty-eight caregivers were presented with a Go-NoGo task to assess inhibitory control. An efficiency score was calculated using their behavioural responses. Eighty-six 6-to-10-month-old infants were presented with a preferential looking task to assess VWM function. VWM load was manipulated across one (low load), two (medium load) and three (high load) items. Functional near-infrared spectroscopy was used to record brain activation from caregivers and their infants. We found no direct association between caregiver efficiency and infant VWM behaviour. However, we found an indirect association - caregiver efficiency was linked to infant VWM through left-lateralized fronto-parietal engagement. Specifically, infants with low efficiency caregivers showed decreasing left-lateralized parietal engagement with increasing VWM performance at the medium and high loads compared to infants with high efficiency caregivers, who did not show any load- or performance-dependent modulation. Our findings contribute to a growing body of literature examining the role that caregivers play in early neurocognitive development.

Stunting in infancy is associated with atypical activation of working memory and attention networks

Stunting is associated with poor long-term cognitive, academic and economic outcomes, yet the mechanisms through which stunting impacts cognition in early development remain unknown. In a first-ever neuroimaging study conducted on infants from rural India, we demonstrate that stunting impacts a critical, early-developing cognitive system-visual working memory. Stunted infants showed poor visual working memory performance and were easily distractible. Poor performance was associated with reduced engagement of the left anterior intraparietal sulcus, a region involved in visual working memory maintenance and greater suppression in the right temporoparietal junction, a region involved in attentional shifting. When assessed one year later, stunted infants had lower problem-solving scores, while infants of normal height with greater left anterior intraparietal sulcus activation showed higher problem-solving scores. Finally, short-for-age infants with poor physical growth indices but good visual working memory performance showed more positive outcomes suggesting that intervention efforts should focus on improving working memory and reducing distractibility in infancy.

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.

A principled link between object naming and representation is available to infants by seven months of age

By their first birthdays, infants represent objects flexibly as a function of not only whether but how the objects are named. Applying the same name to a set of different objects from the same category supports object categorization, with infants encoding commonalities among objects at the expense of individuating details. In contrast, applying a distinct name to each object supports individuation, with infants encoding distinct features at the expense of categorical information. Here, we consider the development of this nuanced link between naming and representation in infants' first year. Infants at 12 months (Study 1; N = 55) and 7 months (Study 2; N = 96) participated in an online recognition memory task. All infants saw the same objects, but their recognition of these objects at test varied as a function of how they had been named. At both ages, infants successfully recognized objects that had been named with distinct labels but failed to recognize these objects when they had all been named with the same, consistent label. This new evidence demonstrates that a principled link between object naming and representation is available by 7 months, early enough to support infants as they begin mapping words to meaning.

Visual Short-Term Memory Persists Across Multiple Fixations: An n-Back Approach to Quantifying Capacity in Infants and Adults

Visual short-term memory (STM) is a foundational component of general cognition that develops rapidly during the first year of life. Although previous research has revealed important relations between overt visual fixation and memory formation, it is unknown whether infants can maintain distinct memories for sequentially fixated items or remember nonfixated array items. Participants (5-month-olds, 11-month-olds, and adults; n = 24 at each age) from the United States were tested in a passive change-detection paradigm with an n-back manipulation to examine memory for the last fixated item (one-back), second-to-last fixated item (two-back), or nonfixated item (change-other). Eye tracking was used to measure overt fixation while participants passively viewed arrays of colored circles. Results for all ages revealed convergent evidence of memory for up to two sequentially fixated objects (i.e., one-back, two-back), with moderate evidence for nonfixated array items (change-other). A permutation analysis examining change preference over time suggested that differences could not be explained by perseverative looking or location biases.

Discovering category boundaries: The role of comparison in infants' novel category learning

A key question in categorization is how infants extract regularities from the exemplars they encounter. Detecting similarities and dissimilarities across items is vital in order to determine category-relevant features. Previous research found evidence that infants acquire a single category more easily with paired presentations in comparison with single presentations (Oakes & Ribar, 2005, Infancy, 7, 85; Oakes & Kovack-Lesh, 2007, Cogniție, Creier, Comportament / Cognition, Brain, Behavior, XI, 661). Here, we focus on infants' acquisition of a category contrast, that is, when they are exposed to two categories. In an eye-tracking study, we examined 10-month-old infants' ability to learn two novel visual categories when presented with one item at a time and with items in pairs. Infants were familiarized with pairs of items from the same category or with pairs of items from different categories (cross-category pairs). Using a linear model with a priori contrasts, we show that infants' learning is directly related to the opportunity for category comparison: There is no evidence of category learning in the single-item condition, improved performance when familiarized with same-category pairs, and finally robust category learning when familiarized with cross-category pairs. We conclude that comparison which involves items from different categories promotes category formation, by highlighting differences and promoting a discovery of category boundaries.

The role of redundant verbal labels in 8- and 10-month-olds' working memory

Verbal labels have been shown to help preverbal infants' performance on various cognitive tasks, such as categorization. Redundant labels also aid adults' visual working memory (WM), but it is not known if this linguistic benefit extends to preverbal infants' WM. In two eye-tracking studies, we tested whether 8- and 10-month-old infants' WM performance would improve with the presence of redundant labels in a Delayed Match Retrieval (DMR) paradigm that tested infants' WM for object-location bindings. Findings demonstrated that infants at both ages were unable to remember two object-location bindings when co-presented with labels at encoding. Moreover, infants who encoded the object-location bindings with labels were not significantly better than those who did so in silence. These findings are discussed in the context of label advantages in cognition and auditory dominance.

Attentional blink in preverbal infants

Primary cognitive processes, such as spatial attention, are essential to our higher cognitive abilities and develop dramatically in the first year of life. The spatial aspect of infants' working memory is equivalent to that of adults. However, it is unclear whether this is true for the temporal domain. Thus, we investigated the temporal aspect of infants' working memory using an attentionally demanding task by focusing on the attentional blink effect, in which the identification of the second of the two brief targets is impaired when inter-target lags are short. We argue that finding a similar pattern of the attentional blink in preverbal infants and adults indicates that infants can complete the consolidation of the first target into working memory at a similar temporal scale as adults. In this experiment, we presented 7- to 8-month-old infants with rapid serial visual streams at a rate of 100 ms/item, including two female faces as targets, and examined whether they could identify the targets by measuring their preference to novel faces compared to targets. The temporal separation between the two targets was 200 or 800 ms. We found that the infants could identify both targets under the longer lag, but they failed to identify the second target under the shorter lag. The adult experiment using the same temporal separation as in the infant experiment revealed the attentional blink effect. These results suggest that 7- to 8-month-old infants can consolidate two items into working memory by 800 ms but not by 200 ms.

Association of Retinal Microvascular Characteristics With Short-term Memory Performance in Children Aged 4 to 5 Years

IMPORTANCE

Neurocognitive functions develop rapidly in early childhood and depend on the intrinsic cooperation between cerebral structures and the circulatory system. The retinal microvasculature can be regarded as a mirror image of the cerebrovascular circulation.

OBJECTIVE

To investigate the association between retinal vessel characteristics and neurological functioning in children aged 4 to 5 years.

DESIGN, SETTING, AND PARTICIPANTS

In this cohort study, mother-child pairs were recruited at birth from February 10, 2010, to June 24, 2014, and renewed consent at their follow-up visit from December 10, 2014, to July 13, 2018. Participants were followed up longitudinally within the prospective Environmental Influence on Aging in Early Life birth cohort. A total of 251 children underwent assessment for this study. Data were analyzed from July 17 to October 30, 2019.

MAIN OUTCOMES AND MEASURES

Retinal vascular diameters, the central retinal arteriolar equivalent (CRAE), central retinal venular equivalent (CRVE), vessel tortuosity, and fractal dimensions were determined. Attention and psychomotor speed, visuospatial working memory, and short-term visual recognition memory were assessed by the Cambridge Neuropsychological Test Automated Battery, including the following tasks: Motor Screening (MOT), Big/Little Circle (BLC), Spatial Span (SSP), and Delayed Matching to Sample (DMS).

RESULTS

Among the 251 children included in the assessment (135 girls [53.8%]; mean [SD] age, 4.5 [0.4] years), for every 1-SD widening in CRVE, the children performed relatively 2.74% (95% CI, -0.12 to 5.49; P = .06) slower on the MOT test, had 1.76% (95% CI, -3.53% to -0.04%; P = .04) fewer correct DMS assessments in total, and made 2.94% (95% CI, 0.39 to 5.29; P = .02) more errors given a previous correct answer in the DMS task on multiple linear regression modeling. For every 1-SD widening in CRAE, the total percentage of errors and errors given previous correct answers in the DMS task increased 1.44% (95% CI, -3.25% to 0.29%; P = .09) and 2.30% (95% CI, -0.14% to 4.61%; P = .07), respectively. A 1-SD higher vessel tortuosity showed a 4.32% relative increase in latency in DMS task performance (95% CI, -0.48% to 9.12%; P = .07). Retinal vessel characteristics were not associated with BLC and SSP test outcomes.

CONCLUSIONS AND RELEVANCE

These findings suggest that children's microvascular phenotypes are associated with short-term memory and that changes in the retinal microvasculature may reflect neurological development during early childhood.

The functional brain networks that underlie visual working memory in the first two years of life

Visual working memory (VWM) is a central cognitive system used to compare views of the world and detect changes in the local environment. This system undergoes dramatic development in the first two years; however, we know relatively little about the functional organization of VWM at the level of the brain. Here, we used image-based functional near-infrared spectroscopy (fNIRS) to test four hypotheses about the spatial organization of the VWM network in early development. Four-month-olds, 1-year-olds, and 2-year-olds completed a VWM task while we recorded neural activity from 19 cortical regions-of-interest identified from a meta-analysis of the adult fMRI literature on VWM. Results showed significant task-specific functional activation near 6 of 19 ROIs, revealing spatial consistency in the brain regions activated in our study and brain regions identified to be part of the VWM network in adult fMRI studies. Working memory related activation was centered on bilateral anterior intraparietal sulcus (aIPS), left temporoparietal junction (TPJ), and left ventral occipital complex (VOC), while visual exploratory measures were associated with activation in right dorsolateral prefrontal cortex, left TPJ, and bilateral IPS. Results show that a distributed brain network underlies functional changes in VWM in infancy, revealing new insights into the neural mechanisms that support infants’ improved ability to remember visual information and to detect changes in an on-going visual stream.

•

A distributed brain network underlies functional changes in VWM in infancy and toddlerhood.

•

This network shows robust engagement of similar brain regions identified in fMRI studies with adults as early as four months.

•

Working memory related activation was centered on bilateral anterior intraparietal sulcus, left temporoparietal junction, and left ventral occipital complex

•

Visual exploratory measures were associated with activation in right dorsolateral prefrontal cortex, bilateral anterior intraparietal sulcus, and left temporoparietal junction.

Robust data and power in infant research: A case study of the effect of number of infants and number of trials in visual preference procedures

As in many areas of science, infant research suffers from low power. The problem is further compounded in infant research because of the difficulty in recruiting and testing large numbers of infant participants. Researchers have been searching for a solution and, as illustrated by this special section, have been focused on getting the most out of infant data. We illustrate one solution by showing how we can increase power in visual preference tasks by increasing the amount of data obtained from each infant. We discuss issues of power and present work examining how, under some circumstances, power is increased by increasing the precision of measurement. We report the results of a series of simulations based on a sample of visual preference task data collected from three infant laboratories showing how more powerful research designs can be achieved by including more trials per infant. Implications for infant procedures in general are discussed.

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.

Persistence and Accumulation of Visual Memories for Objects in Scenes in 12-Month-Old Infants

Visual memory for objects has been studied extensively in infants over the past 20 years, however, little is known about how they are formed when objects are embedded in naturalistic scenes. In adults, memory for objects in a scene show information accumulation over time as well as persistence despite interruptions (Melcher, 2001, 2006). In the present study, eye-tracking was used to investigate these two processes in 12-month-old infants (N = 19) measuring: (1) whether longer encoding time can improve memory performance (accumulation), and (2) whether multiple shorter exposures to a scene are equivalent to a single exposure of the same total duration (persistence). A control group of adults was also tested in a closely matched paradigm (N = 23). We found that increasing exposure time led to gains in memory performance in both groups. Infants were found to be successful in remembering objects with continuous exposures to a scene, but unlike adults, were not able to perform better than chance when interrupted. However, infants' scan patterns showed evidence of memory as they continued the exploration of the scene in a strategic way following the interruption. Our findings provide insight into how infants are able to build representations of their visual environment by accumulating information about objects embedded in scenes.

Two-year-olds succeed at MIT: Multiple identity tracking in 20- and 25-month-old infants

Infants' ability to remember objects and their locations emerges during the first year of life. However, not much is known about infants' ability to track objects' identities in a dynamic environment. Here, we tailored the delayed match retrieval eye-tracking paradigm to study infants' ability to track two object identities during occlusion-an infant version of multiple identity tracking (MIT). Delayed match retrieval uses virtual "cards" as stimuli that are first shown face up, exposing to-be-remembered information, and then turned face down, occluding it. Here, cards were subject to movement during the face-down occlusion period. We used complex non-nameable objects as card faces to discourage verbal rehearsal. In three experiments (N = 110), we compared infants' ability to track object identities when two previously exposed cards were static (Experiment 1), were moved into new positions along the same trajectory (Experiment 2), or were moved along different trajectories (Experiment 3) while face down. We found that 20-month-olds could remember two object identities when static; however, it was not until 25 months of age that infants could track when movement was introduced. Our results show that the ability to track multiple identities in visual working memory is present by 25 months.

When learning goes beyond statistics: Infants represent visual sequences in terms of chunks

Much research has documented infants' sensitivity to statistical regularities in auditory and visual inputs, however the manner in which infants process and represent statistically defined information remains unclear. Two types of models have been proposed to account for this sensitivity: statistical models, which posit that learners represent statistical relations between elements in the input; and chunking models, which posit that learners represent statistically-coherent units of information from the input. Here, we evaluated the fit of these two types of models to behavioral data that we obtained from 8-month-old infants across four visual sequence-learning experiments. Experiments examined infants' representations of two types of structures about which statistical and chunking models make contrasting predictions: illusory sequences (Experiment 1) and embedded sequences (Experiments 2-4). In all four experiments, infants discriminated between high probability sequences and low probability part-sequences, providing strong evidence of learning. Critically, infants also discriminated between high probability sequences and statistically-matched sequences (illusory sequences in Experiment 1, embedded sequences in Experiments 2-3), suggesting that infants learned coherent chunks of elements. Experiment 4 examined the temporal nature of chunking, and demonstrated that the fate of embedded chunks depends on amount of exposure. These studies contribute important new data on infants' visual statistical learning ability, and suggest that the representations that result from infants' visual statistical learning are best captured by chunking models.

Visual working memory in early development: a developmental cognitive neuroscience perspective

In this article, we review the literature on the development of visual working memory (VWM). We focus on two major periods of development, infancy and early childhood. First, we discuss the innovative methods that have been devised to understand how the development of selective attention and perception provide the foundation of VWM abilities. We detail the behavioral and neural data associated with the development of VWM during infancy. Next, we discuss various signatures of development in VWM during early childhood in the context of spatial and featural memory processes. We focus on the developmental transition to more adult-like VWM properties. Finally, we discuss computational frameworks that have explained the complex patterns of behavior observed in VWM tasks from infancy to adulthood and attempt to explain links between measures of infant VWM and childhood VWM.

The development of object-based attention in infants

We examined whether infants aged 6-8 months show object-based attention using the preferential looking method. In object-based attention, which is a prerequisite function for efficient real-world processing of various stimuli, a target that appears at a cued object is detected and processed faster than a target appearing at an uncued object. We presented 6- to 8-month-old infants with the visual stimuli consisting of two white vertical rectangles side by side, in which a target appearing at 1) the cued location, 2) the end opposite to the cued location, and 3) another rectangle's end following the cue, using an established paradigm, and measured each infant's first saccade to the target. We found that (1) infants of all ages could make the first saccade to the target appearing at the cued location, (2) only 8-month-old infants made the first saccade to the target appearing at the opposite end to the cued location more accurately than to the target appearing at the other rectangle's end. These results indicate that object-based attention might be acquired around 8 months compared with the spatial cueing effect that appears at around 6 months. Our findings suggest that the objects play a role in visual attention in 8-month-old infants.

10-Month-Old Infants Are Sensitive to the Time Course of Perceived Actions: Eye-Tracking and EEG Evidence

Research has shown that infants are able to track a moving target efficiently - even if it is transiently occluded from sight. This basic ability allows prediction of when and where events happen in everyday life. Yet, it is unclear whether, and how, infants internally represent the time course of ongoing movements to derive predictions. In this study, 10-month-old crawlers observed the video of a same-aged crawling baby that was transiently occluded and reappeared in either a temporally continuous or non-continuous manner (i.e., delayed by 500 ms vs. forwarded by 500 ms relative to the real-time movement). Eye movement and rhythmic neural brain activity (EEG) were measured simultaneously. Eye movement analyses showed that infants were sensitive to slight temporal shifts in movement continuation after occlusion. Furthermore, brain activity associated with sensorimotor processing differed between observation of continuous and non-continuous movements. Early sensitivity to an action's timing may hence be explained within the internal real-time simulation account of action observation. Overall, the results support the hypothesis that 10-month-old infants are well prepared for internal representation of the time course of observed movements that are within the infants' current motor repertoire.

Sample size, statistical power, and false conclusions in infant looking-time research

Infant research is hard. It is difficult, expensive, and time consuming to identify, recruit and test infants. As a result, ours is a field of small sample sizes. Many studies using infant looking time as a measure have samples of 8 to 12 infants per cell, and studies with more than 24 infants per cell are uncommon. This paper examines the effect of such sample sizes on statistical power and the conclusions drawn from infant looking time research. An examination of the state of the current literature suggests that most published looking time studies have low power, which leads in the long run to an increase in both false positive and false negative results. Three data sets with large samples (>30 infants) were used to simulate experiments with smaller sample sizes; 1000 random subsamples of 8, 12, 16, 20, and 24 infants from the overall samples were selected, making it possible to examine the systematic effect of sample size on the results. This approach revealed that despite clear results with the original large samples, the results with smaller subsamples were highly variable, yielding both false positive and false negative outcomes. Finally, a number of emerging possible solutions are discussed.

Real-world visual statistics and infants' first-learned object names

We offer a new solution to the unsolved problem of how infants break into word learning based on the visual statistics of everyday infant-perspective scenes. Images from head camera video captured by 8 1/2 to 10 1/2 month-old infants at 147 at-home mealtime events were analysed for the objects in view. The images were found to be highly cluttered with many different objects in view. However, the frequency distribution of object categories was extremely right skewed such that a very small set of objects was pervasively present-a fact that may substantially reduce the problem of referential ambiguity. The statistical structure of objects in these infant egocentric scenes differs markedly from that in the training sets used in computational models and in experiments on statistical word-referent learning. Therefore, the results also indicate a need to re-examine current explanations of how infants break into word learning.This article is part of the themed issue 'New frontiers for statistical learning in the cognitive sciences'.

Delayed Match Retrieval: a novel anticipation-based visual working memory paradigm

We tested 8- and 10-month-old infants' visual working memory (VWM) for object-location bindings - what is where - with a novel paradigm, Delayed Match Retrieval, that measured infants' anticipatory gaze responses (using a Tobii T120 eye tracker). In an inversion of Delayed-Match-to-Sample tasks and with inspiration from the game Memory, in test trials, three face-down virtual 'cards' were presented. Two flipped over sequentially (revealing, e.g. a swirl pattern and then a star), and then flipped back face-down. Next, the third card was flipped to reveal a match (e.g. a star) to one of the previously seen, now face-down cards. If infants looked to the location where the (now face-down) matching card had been shown, this was coded as a correct response. To encourage anticipatory looks, infants subsequently received a reward (a brief, engaging animation) presented at that location. Ten-month-old infants performed significantly above chance, showing that their VWM could hold object-location information for the two cards. Overall, 8-month-olds' performance was at chance, but they showed a robust learning trend. These results corroborate previous findings (Kaldy & Leslie, 2005; Oakes, Ross-Sheehy & Luck, 2006) and point to rapid development of VWM for object-location bindings. However, compared to previous paradigms that measure passive gaze responses to novelty, this paradigm presents a more challenging, ecologically relevant test of VWM, as it measures the ability to make online predictions and actively localize objects based on VWM. In addition, this paradigm can be readily scaled up to test toddlers or older children without significant modification.

Varieties of Visual Working Memory Representation in Infancy and Beyond

Research on the developmental origins of visual working memory in infants has largely progressed along two separate branches. One branch is rooted in the classic work on adult visual working memory, while the other is rooted in the classic work on the object concept in infancy. Both lines of research have yielded some converging results but also some surprisingly different patterns. In this review, I show that these different patterns are evidence for two distinct types of representations, which I term feature-based and object-based. I then show that there is evidence for both representation types beyond infancy.

Visual working memory continues to develop through adolescence

The capacity of visual working memory (VWM) refers to the amount of visual information that can be maintained in mind at once, readily accessible for ongoing tasks. In healthy young adults, the capacity limit of VWM corresponds to about three simple objects. While some researchers argued that VWM capacity becomes adult-like in early years of life, others claimed that the capacity of VWM continues to develop beyond middle childhood. Here we assessed whether VWM capacity reaches adult levels in adolescence. Using an adaptation of the visual change detection task, we measured VWM capacity estimates in 13-year-olds, 16-year-olds, and young adults. We tested whether the capacity estimates observed in early or later years of adolescence were comparable to the estimates obtained from adults. Our results demonstrated that the capacity of VWM continues to develop throughout adolescence, not reaching adult levels even in 16-year-olds. These findings suggest that VWM capacity displays a prolonged development, similar to the protracted trajectories observed in various other aspects of cognition.

Array heterogeneity prevents catastrophic forgetting in infants

Working memory is limited in adults and infants. But unlike adults, infants whose working memory capacity is exceeded often fail in a particularly striking way: they do not represent any of the presented objects, rather than simply remembering as many objects as they can and ignoring anything further (Feigenson & Carey, 2003, 2005). Here we explored the nature of this "catastrophic forgetting," asking whether stimuli themselves modulate the way in which infants' memory fails. We showed 13-month old infants object arrays that either were within or that exceeded working memory capacity--but, unlike previous experiments, presented objects with contrasting features. Although previous studies have repeatedly documented infants' failure to represent four identical hidden objects, in Experiments 1 and 2 we found that infants who saw four contrasting objects hidden, and then retrieved just two of the four, successfully continued searching for the missing objects. Perceptual contrast between objects sufficed to drive this success; infants succeeded regardless of whether the different objects were contrastively labeled, and regardless of whether the objects were semantically familiar or completely novel. In Experiment 3 we explored the nature of this surprising success, asking whether array heterogeneity actually expanded infants' working memory capacity or rather prevented catastrophic forgetting. We found that infants successfully continued searching after seeing four contrasting objects hidden and retrieving two of them, but not after retrieving three of them. This suggests that, like adults, infants were able to remember up to, but not beyond, the limits of their working memory capacity when representing heterogeneous arrays.