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.     

Mechanisms of Categorization in Infancy

This article presents a connectionist model of correlation‐based categorization by 10‐month‐old infants (Younger, 1985). Simple autoencoder networks were exposed to the same stimuli used to test 10‐month‐olds. The familiarization regime was kept as close as possible to that used with the infants. The performance of the model matched that of the infants. Both infants and networks used covariation information (when available) to segregate items into separate categories. The model provides a mechanistic account of category learning within a test session. It demonstrates how categorization arises as the product of an inextricable interaction between the subject (the infant) and the environment (the stimuli). The computational characteristics of both subject and environment must be considered in conjunction to understand the observed behaviors.     

Learned Labels Shape Pre‐speech Infants’ Object Representations

Infants rapidly learn both linguistic and nonlinguistic representations of their environment and begin to link these from around 6 months. While there is an increasing body of evidence for the effect of labels heard in‐task on infants’ online processing, whether infants’ learned linguistic representations shape learned nonlinguistic representations is unclear. In this study 10‐month‐old infants were trained over the course of a week with two 3D objects, one labeled, and one unlabeled. Infants then took part in a looking time task in which 2D images of the objects were presented individually in a silent familiarization phase, followed by a preferential looking trial. During the critical familiarization phase, infants looked for longer at the previously labeled stimulus than the unlabeled stimulus, suggesting that learning a label for an object had shaped infants’ representations as indexed by looking times. We interpret these results in terms of label activation and novelty response accounts and discuss implications for our understanding of early representational development.     

Infants Discriminate the Affective Expressions of their Peers: The Roles of Age and Familiarization Time

Research examining infants’ discrimination of affect often uses unfamiliar faces and voices of adults. Recently, research has examined infant discrimination of affect in familiar faces and voices. In much of this research, infants were habituated to the affective expressions using a “standard” 50% habituation criterion. We extend this line of research by examining infants’ discrimination of unfamiliar peers’, that is, 4‐month‐olds, dynamic, facial, and vocal affective expressions and assessing how discrimination is affected by changing the habituation criterion. In two experiments, using an infant‐controlled habituation design, we explored 3‐ and 5‐month‐olds’ discrimination of their peers’ dynamic audiovisual displays of positive and negative expressions of affect. Results of Experiment 1, using a 50% habituation criterion, revealed that 5‐month‐olds, but not 3‐month‐olds discriminated the affective expressions of their peers. In Experiment 2, we examined whether 3‐month‐olds’ lack of discrimination in Experiment 1 was a result of insufficient habituation (i.e., familiarization). Specifically, 3‐month‐olds were habituated using a 70% habituation criterion, providing them with longer familiarization time. Results revealed that using the more stringent habituation criterion, 3‐month‐olds showed longer habituation times, that is increased familiarization, and discriminated their peers’ affective expressions. Results are discussed in terms of infants’ discrimination of affect, the role of familiarization time, and limitations of the 50% habituation criterion.     

5‐Month‐Olds’ Categorization of Novel Objects: Task and Measure Dependence

To examine key parameters of the initial conditions in early category learning, two studies compared 5‐month‐olds’ object categorization between tasks involving previously unseen novel objects, and between measures within tasks. Infants in Experiment 1 participated in a visual familiarization–novelty preference (VFNP) task with two‐dimensional (2D) stimulus images. Infants provided no evidence of categorization by either their looking or their examining even though infants in previous research systematically categorized the same objects by examining when they could handle them directly. Infants in Experiment 2 participated in a VFNP task with 3D stimulus objects that allowed visual examination of objects’ 3D instantiation while denying manual contact with the objects. Under these conditions, infants demonstrated categorization by examining but not by looking. Focused examination appears to be a key component of young infants’ ability to form category representations of novel objects, and 3D instantiation appears to better engage such examining.     

Infants' Looking at Possible and Impossible Screen Rotations: The Role of Familiarization

The effects of familiarization and age were examined using Baillargeon's rotating screen paradigm. In Condition A, 4‐month‐olds exposed to 7 180° familiarization trials looked significantly longer at the 180° test events than the 112° test events; there was a familiarity preference. In Condition B, which consisted of 12 instead of 7 180° familiarization trials, the 4‐month‐olds looked significantly longer at the 112° test events than the 180° test events; there was a novelty preference. In Condition C, which was similar to Condition A except that there were 112° familiarization trials, the infants showed a familiarity preference. Thus, 4‐month‐olds experiencing 7 familiarization trials exhibited a familiarity preference, and those experiencing 12 familiarization trials showed a novelty preference. In Condition D, 6‐month‐olds experienced 7 180° familiarization trials; there was no preferential looking to either familiar (180°) test events or novel (112°) test events. Therefore, looking behavior during the test trials was a function of the type of familiarization experience and age and not necessarily an inferred violation of physics.     

Learning Stimulus‐Location Associations in 8‐ and 11‐Month‐Old Infants: Multimodal Versus Unimodal Information

Research on the influence of multimodal information on infants' learning is inconclusive. While one line of research finds that multimodal input has a negative effect on learning, another finds positive effects. The present study aims to shed some new light on this discussion by studying the influence of multimodal information and accompanying stimulus complexity on the learning process. We assessed the influence of multimodal input on the trial‐by‐trial learning of 8‐ and 11‐month‐old infants. Using an anticipatory eye movement paradigm, we measured how infants learn to anticipate the correct stimulus–location associations when exposed to visual‐only, auditory‐only (unimodal), or auditory and visual (multimodal) information. Our results show that infants in both the multimodal and visual‐only conditions learned the stimulus–location associations. Although infants in the visual‐only condition appeared to learn in fewer trials, infants in the multimodal condition showed better anticipating behavior: as a group, they had a higher chance of anticipating correctly on more consecutive trials than infants in the visual‐only condition. These findings suggest that effects of multimodal information on infant learning operate chiefly through effects on infants' attention.     

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.     

Infants’ Discrimination of Familiar and Unfamiliar Accents in Speech

This study investigates infants’ discrimination abilities for familiar and unfamiliar regional English accents. Using a variation of the head‐turn preference procedure, 5‐month‐old infants demonstrated that they were able to distinguish between their own South‐West English accent and an unfamiliar Welsh English accent. However, this distinction was not seen when two unfamiliar accents (Welsh English and Scottish English) were presented to the infants, indicating they had not acquired the general ability to distinguish between regional varieties, but only the distinction between their home accent and unfamiliar regional variations. This ability was also confirmed with 7‐month‐olds, challenging recent claims that infants lose their sensitivity to dialects at around that age. Taken together, our results argue in favor of an early sensitivity to the intonation system of languages, and to the early learning of accent‐specific intonation and potentially segmental patterns. Implications for the development of accent normalization abilities are discussed.     

A Comparison of Infants' Categorization in Paired and Successive Presentation Familiarization Tasks

Three experiments directly compared infants' categorization in variations of the visual familiarization task. In each experiment, 4‐ or 6‐month‐old infants were familiarized with a collection of dogs or cats and then their response to novel dogs and cats was assessed. In Experiment 1, 4‐month‐old infants responded to the exclusive distinction of dogs or cats when tested in a paired‐comparison task. In Experiments 2 and 3, 6‐month‐old infants, but not 4‐month‐old infants, responded to this same distinction in a successive presentation task, even when the amount of familiarization was equated to that of the paired comparison task. Therefore, familiarization with a particular set of stimuli does not induce infants to respond to a single category but rather they respond to different categories depending on features of the task.     

Object Function and Categorization in Infancy: Two Mechanisms of Facilitation

Does function facilitate categorization by focusing infants' attention generally on all commonalities among objects or specifically on functionally relevant properties? After familiarization to a novel category, 18‐month‐olds selected another category member from a pair of previously unseen test objects. In Experiments 1 and 2, infants chose globally similar over functionally similar and novel test objects. Functionally similar and novel test objects were chosen equally. These data suggest that function facilitates categorization through a general attention‐enhancing mechanism. However, when functions were more uniquely and transparently tied to object properties in Experiments 3 and 4, infants chose functionally similar over novel test objects. Globally and functionally similar test objects were chosen equally. Therefore, a specific attention‐enhancing mechanism also sometimes supports categorization.     

Global‐Before‐Basic Object Categorization in Connectionist Networks and 2‐Month‐Old Infants

A 3‐layered backpropagation connectionist network, configured as an autoassociator, learned to form global (e.g., mammal) before basic‐level (e.g., cat) category representations from perceptual input. To test the predicted global‐to‐basic order of category learning of the network, 2‐month‐olds were administered the familiarization/novelty‐preference procedure and examined for representation of global and basic‐level categories. Infants formed a global category representation for mammals that excluded furniture but not a basic‐level representation for cats that excluded elephants, rabbits, or dogs. The empirical results are consistent with the global‐to‐basic learning sequence observed in the network simulations.     

Neural Correlates of Individual Differences in Infant Visual Attention and Recognition Memory

Past studies have identified individual differences in infant visual attention based upon peak look duration during initial exposure to a stimulus. Colombo and colleagues found that infants that demonstrate brief visual fixations (i.e., short lookers) during familiarization are more likely to demonstrate evidence of recognition memory during subsequent stimulus exposure than infants that demonstrate long visual fixations (i.e., long lookers). This study utilized event‐related potentials (ERPs) to examine possible neural mechanisms associated with individual differences in visual attention and recognition memory for 6‐ and 7.5‐month‐old infants. Short‐ and long‐looking infants viewed images of familiar and novel objects during ERP testing. There was a stimulus type by looker type interaction at temporal and frontal electrodes on the late slow wave (LSW). Short lookers demonstrated an LSW that was significantly greater in amplitude in response to novel stimulus presentations. No significant differences in LSW amplitude were found based on stimulus type for long lookers. These results indicate deeper processing and recognition memory of the familiar stimulus for short lookers.     

Intersensory Redundancy and 7‐Month‐Old Infants' Memory for Arbitrary Syllable‐Object Relations

Seven‐month‐old infants require redundant information, such as temporal synchrony, to learn arbitrary syllable‐object relations (Gogate & Bahrick, 1998). Infants learned the relations between 2 spoken syllables, /a/ and /i/, and 2 moving objects only when temporal synchrony was present during habituation. This article presents 2 experiments to address infants' memory for these relations. In Experiment 1, infants remembered the syllable‐object relations after 10 min, only when temporal synchrony between the vocalizations and moving objects was provided during learning. In Experiment 2, 7‐month‐olds were habituated to the same syllable‐object pairs in the presence of temporal synchrony and tested for memory after 4 days. Once again, infants learned and showed emerging memory for the syllable‐object relations 4 days after original learning under the temporally synchronous condition. These findings are consistent with the view that prior to symbolic development, infants learn and remember word‐object relations by perceiving redundant information in the vocal and gestural communication of adults.     

Infants' First Words are not Phonetically Specified: Own Name Recognition in British English‐Learning 5‐Month‐Olds

By the end of their first year of life, infants’ representations of familiar words contain phonetic detail; yet little is known about the nature of these representations at the very beginning of word learning. Bouchon et al. ( 2015 ) showed that French‐learning 5‐month‐olds could detect a vowel change in their own name and not a consonant change, but also that infants reacted to the acoustic distance between vowels. Here, we tested British English‐learning 5‐month‐olds in a similar study to examine whether the acoustic/phonological characteristics of the native language shape the nature of the acoustic/phonetic cues that infants pay attention to. In the first experiment, British English‐learning infants failed to recognize their own name compared to a mispronunciation of initial consonant (e.g., Molly versus Nolly) or vowel (e.g., April versus Ipril). Yet in the second experiment, they did so when the contrasted name was phonetically dissimilar (e.g., Sophie versus Amber). Differences in phoneme category (stops versus continuants) between the correct consonant versus the incorrect one significantly predicted infants’ own name recognition in the first experiment. Altogether, these data suggest that infants might enter into a phonetic mode of processing through different paths depending on the acoustic characteristics of their native language.     

Early Word Segmentation in Naturalistic Environments: Limited Effects of Speech Register

We examined 7.5‐month‐old infants' ability to segment words from infant‐ and adult‐directed speech (IDS and ADS). In particular, we extended the standard design of most segmentation studies by including a phase where infants were repeatedly exposed to target word recordings at their own home (extended exposure) in addition to a laboratory‐based familiarization. This enabled us to examine infants' segmentation of words from speech input in their naturalistic environment, extending current findings to learning outside the laboratory. Results of a modified preferential‐listening task show that infants listened longer to isolated tokens of familiarized words from home relative to novel control words regardless of register. However, infants showed no recognition of words exposed to during purely laboratory‐based familiarization. This indicates that infants succeed in retaining words in long‐term memory following extended exposure and recognizing them later on with considerable flexibility. In addition, infants segmented words from both IDS and ADS, suggesting limited effects of speech register on learning from extended exposure in naturalistic environments. Moreover, there was a significant correlation between segmentation success and infants' attention to ADS, but not to IDS, during the extended exposure phase. This finding speaks to current language acquisition models assuming that infants' individual attention to language stimuli drives successful learning.     

Phonotactic Constraints on Infant Word Learning

How do infants use their knowledge of native language sound patterns when learning words? There is ample evidence of infants’ precocious acquisition of native language sound structure during the first year of life, but much less evidence concerning how they apply this knowledge to the task of associating sounds with meanings in word learning. To address this question, 18‐month‐olds were presented with two phonotactically legal object labels (containing sound sequences that occur frequently in English) or two phonotactically illegal object labels (containing sound sequences that never occur in English), paired with novel objects. Infants were then tested using a looking‐while‐listening measure. The results revealed that infants looked at the correct objects after hearing the legal labels, but not the illegal labels. Furthermore, vocabulary size was related to performance. Infants with larger receptive vocabularies displayed greater differences between learning of legal and illegal labels than infants with smaller vocabularies. These findings provide evidence that infants’ knowledge of native language sound patterns influences their word learning.     

A Comparison of Visual Familiarization and Object‐Examining Measures of Categorization in 9‐Month‐Old Infants

Three studies were conducted to determine whether differential patterns of categorization observed in studies using visual familiarization and object‐examining measures hold up as procedural confounds are eliminated. In Experiment 1, we attempted as direct a comparison as possible between visual and object‐examining measures of categorization. Consistent with previous reports, 9‐month‐old infants distinguished a basic‐level contrast (dog–horse) in the visual task, but not in the examining task. Experiment 2 was designed to reduce levels of nonexploratory activity in an examining task; 9‐month‐olds again failed to distinguish categories of dogs and horses. In Experiment 3, we adopted a paired‐comparison test format in the object‐examining task. Infants did display a novel category preference under paired testing conditions. The results suggest that the different patterns of categorization often seen in looking and touching tasks are a reflection, not of different categorization processes, but of the differential sensitivity of the tasks.     

The Cat is out of the Bag: The Joint Influence of Previous Experience and Looking Behavior on Infant Categorization

We examined the effect of 4‐month‐old infants' previous experience with dogs, cats, or both and their online looking behavior on their learning of the adult‐defined category of cat in a visual familiarization task. Four‐month‐old infants' (N = 123) learning in the laboratory was jointly determined by whether or not they had experience with pets at home and how much they shifted their gaze back and forth between the stimuli during familiarization. Specifically, only infants with pets at home who also exhibited high levels of switching during familiarization remembered the individual cat exemplars or formed a summary representation of those cats. These results are consistent with recent theorizing about the processes of how infants' categorical representations are formed, and provide new understanding into how infants' categorization unfolds over time.     

Infants’ Visual Attention to Baby DVDs as a Function of Program Pacing

This study examined the effects of program pacing, defined as the rate of scene and character change per minute, on infants’ visual attention to video presentations. Seventy‐two infants (twenty‐four 6‐month‐olds, twenty‐four 9‐month‐olds, twenty‐four 12‐month‐olds) were exposed to one of two sets of high‐ and low‐paced commercial infant DVDs. Each DVD was approximately 5‐min long, and the order the DVDs were viewed was counterbalanced for pace. Attention was higher during rapidly than slowly paced DVDs, particularly for the 6‐ and 9‐month‐old infants. These results support previous research documenting that attention is initially controlled by exogenous qualities (e.g., rapid pace), but with development and experience becomes more influenced by endogenous factors.