Information from multiple modalities helps 5‐month‐olds learn abstract rules

By 7 months of age, infants are able to learn rules based on the abstract relationships between stimuli ( Marcus et al ., 1999 ), but they are better able to do so when exposed to speech than to some other classes of stimuli. In the current experiments we ask whether multimodal stimulus information will aid younger infants in identifying abstract rules. We habituated 5‐month‐olds to simple abstract patterns (ABA or ABB) instantiated in coordinated looming visual shapes and speech sounds (Experiment 1), shapes alone (Experiment 2), and speech sounds accompanied by uninformative but coordinated shapes (Experiment 3). Infants showed evidence of rule learning only in the presence of the informative multimodal cues. We hypothesize that the additional evidence present in these multimodal displays was responsible for the success of younger infants in learning rules, congruent with both a Bayesian account and with the Intersensory Redundancy Hypothesis.     

Enhanced music sensitivity in 9-month-old bilingual infants

This study explores the influence of bilingualism on the cognitive processing of language and music. Specifically, we investigate how infants learning a non-tone language perceive linguistic and musical pitch and how bilingualism affects cross-domain pitch perception. Dutch monolingual and bilingual infants of 8–9 months participated in the study. All infants had Dutch as one of the first languages. The other first languages, varying among bilingual families, were not tone or pitch accent languages. In two experiments, infants were tested on the discrimination of a lexical (N = 42) or a violin (N = 48) pitch contrast via a visual habituation paradigm. The two contrasts shared identical pitch contours but differed in timbre. Non-tone language learning infants did not discriminate the lexical contrast regardless of their ambient language environment. When perceiving the violin contrast, bilingual but not monolingual infants demonstrated robust discrimination. We attribute bilingual infants’ heightened sensitivity in the musical domain to the enhanced acoustic sensitivity stemming from a bilingual environment. The distinct perceptual patterns between language and music and the influence of acoustic salience on perception suggest processing diversion and association in the first year of life. Results indicate that the perception of music may entail both shared neural network with language processing, and unique neural network that is distinct from other cognitive functions.     

Distinguishing novelty and familiarity effects in infant preference procedures

This paper considers possible problems researchers might face when interpreting the results of studies that employ variants of the preference procedure. Infants show a tendency to shift their preference from familiar to novel stimuli with increasing exposure to the familiar stimulus, a behaviour that is exploited by the habituation paradigm. This change in attentional preference with exposure leads us to suggest that researchers interested in infants' pre‐experimental or spontaneous preferences should beware of the potentially confounding effects of exposing infants to familiarization trials prior to employing the preference procedure. The notion that infant attentional preference is dynamic also calls into question the use of the direction of post‐familiarization preference per se when interpreting the knowledge or strategies available to infants. We look into the results of a cross‐modal word learning study to show how the interpretation of results may be difficult when infants exhibit a significant preference in an unexpected direction. As a possible solution to this problem we propose that significant preferences in both directions should be sought at multiple intervals over time. Copyright © 2004 John Wiley & Sons, Ltd.     

Linking patterns of infant eye movements to a neural network model of the ventral stream using representational similarity analysis

Little is known about the development of higher-level areas of visual cortex during infancy, and even less is known about how the development of visually guided behavior is related to the different levels of the cortical processing hierarchy. As a first step toward filling these gaps, we used representational similarity analysis (RSA) to assess links between gaze patterns and a neural network model that captures key properties of the ventral visual processing stream. We recorded the eye movements of 4- to 12-month-old infants (N = 54) as they viewed photographs of scenes. For each infant, we calculated the similarity of the gaze patterns for each pair of photographs. We also analyzed the images using a convolutional neural network model in which the successive layers correspond approximately to the sequence of areas along the ventral stream. For each layer of the network, we calculated the similarity of the activation patterns for each pair of photographs, which was then compared with the infant gaze data. We found that the network layers corresponding to lower-level areas of visual cortex accounted for gaze patterns better in younger infants than in older infants, whereas the network layers corresponding to higher-level areas of visual cortex accounted for gaze patterns better in older infants than in younger infants. Thus, between 4 and 12 months, gaze becomes increasingly controlled by more abstract, higher-level representations. These results also demonstrate the feasibility of using RSA to link infant gaze behavior to neural network models. A video abstract of this article can be viewed at https://youtu.be/K5mF2Rw98Is     

Only the right noise? Effects of phonetic and visual input variability on 14‐month‐olds' minimal pair word learning

Seminal work by Werker and colleagues (Stager & Werker [1997] Nature, 388, 381–382) has found that 14‐month‐old infants do not show evidence for learning minimal pairs in the habituation‐switch paradigm. However, when multiple speakers produce the minimal pair in acoustically variable ways, infants' performance improves in comparison to a single speaker condition (Rost & McMurray [2009] Developmental Science, 12, 339–349). The current study further extends these results and assesses how different kinds of input variability affect 14‐month‐olds' minimal pair learning in the habituation‐switch paradigm testing German learning infants. The first two experiments investigated word learning when the labels were spoken by a single speaker versus when the labels were spoken by multiple speakers. In the third experiment we studied whether non‐acoustic variability, implemented by visual variability of the objects presented together with the labels, would also affect minimal pair learning. We found enhanced learning in the multiple speakers compared to the single speaker condition, confirming previous findings with English‐learning infants. In contrast, visual variability of the presented objects did not support learning. These findings both confirm and better delimit the beneficial role of speech‐specific variability in minimal pair learning. Finally, we review different proposals on the mechanisms via which variability confers benefits to learning and outline what may be likely principles that underlie this benefit. We highlight among these the multiplicity of acoustic cues signalling phonemic contrasts and the presence of relations among these cues. It is in these relations where we trace part of the source for the apparent paradoxical benefit of variability in learning.     

Robustness of the rule-learning effect in 7-month-old infants: A close,multicenter replication of Marcus et al. (1999)

We conducted a close replication of the seminal work by Marcus and colleagues from 1999, which showed that after a brief auditory exposure phase, 7-month-old infants were able to learn and generalize a rule to novel syllables not previously present in the exposure phase. This work became the foundation for the theoretical framework by which we assume that infants are able to learn abstract representations and generalize linguistic rules. While some extensions on the original work have shown evidence of rule learning, the outcomes are mixed, and an exact replication of Marcus et al.'s study has thus far not been reported. A recent meta-analysis by Rabagliati and colleagues brings to light that the rule-learning effect depends on stimulus type (e.g., meaningfulness, speech vs. nonspeech) and is not as robust as often assumed. In light of the theoretical importance of the issue at stake, it is appropriate and necessary to assess the replicability and robustness of Marcus et al.'s findings. Here we have undertaken a replication across four labs with a large sample of 7-month-old infants (N = 96), using the same exposure patterns (ABA and ABB), methodology (Headturn Preference Paradigm), and original stimuli. As in the original study, we tested the hypothesis that infants are able to learn abstract “algebraic” rules and apply them to novel input. Our results did not replicate the original findings: infants showed no difference in looking time between test patterns consistent or inconsistent with the familiarization pattern they were exposed to.     

Categorization and understanding of facial expressions in 4‐month‐old infants

This study examined whether 4‐month‐olds (N = 40) could perceptually categorize happy and angry faces, and show appropriate behavior in response to these faces. During the habituation phase, infants were shown the same type of facial expressions (happy or angry) posed by three models, and their behavior in response to those faces was observed. During the test phase immediately after the habituation phase, infants saw a novel emotional expression and a familiar expression posed by a new model, and their looking times were measured. The results indicated that, although 4‐month‐olds could perceptually categorize happy and angry faces accurately, they responded positively to both expression types. These findings suggest that, although infants can perceptually categorize facial expressions at 4 months of age, they require further time to learn the affective meanings of the facial expressions.     

Infant visual habituation

The use of visual habituation in the study of infant cognition and learning is reviewed. This article traces the history of the technique, underlying theory, and procedural variation in its measurement. In addition, we review empirical findings with respect to the cognitive processes that presumably contribute to habituation, studies of developmental course and long-term prediction, as well as recent attempts to address or explain the phenomenon of visual habituation through the use of mathematical or quantitative models. The review ends with an appeal for a return to the study of habituation per se as a valid measure of infant learning, rather than relegating the phenomenon to its use as a technique for familiarizing infants in procedures testing for discrimination or recognition.     

The link between infant attention to goal‐directed action and later theory of mind abilities

Various studies have shown that infants in their first year of life are able to interpret human actions as goal‐directed. It is argued that this understanding is a precondition for understanding intentional actions and attributing mental states. Moreover, some authors claim that this early action understanding is a precursor of later Theory of Mind (ToM) development. To test this, we related 6‐month‐olds’ performance in an action interpretation task to their performance in ToM tasks at the age of 4 years. Action understanding was assessed using a modified version of the Woodward‐paradigm ( Woodward, 1999 ). At the age of 4 years, the same children were tested with the German version of the ToM scale developed by Wellman and Liu (2004 ). Results revealed a correlation between infants’ decrement of attention to goal‐directed action and their ability to solve a false belief task at the age of 4 years with no modulation by language abilities. Our results indicate a link between infant attention to goal‐directed action and later theory of mind abilities.     

Learning representations of wordforms with recurrent networks: comment on sibley, kello, plaut, & elman (2008)

Sibley et al. (2008) report a recurrent neural network model designed to learn wordform representations suitable for written and spoken word identification. The authors claim that their sequence encoder network overcomes a key limitation associated with models that code letters by position (e.g., CAT might be coded as C‐in‐position‐1, A‐in‐position‐2, T‐in‐position‐3). The problem with coding letters by position (slot‐coding) is that it is difficult to generalize knowledge across positions; for example, the overlap between CAT and TOMCAT is lost. Although we agree this is a critical problem with many slot‐coding schemes, we question whether the sequence encoder model addresses this limitation, and we highlight another deficiency of the model. We conclude that alternative theories are more promising.     

Embodied infant attention

Does real time coupling between mental and physical activity early in development have functional significance? To address this question, we examined the habituation of visual attention and the subsequent response to change in two groups of 3‐month‐olds with different patterns of movement–attention coupling. In suppressors, the typical decrease in body movement at the onset of looks persists into the looks. In rebounders, the initial decrease is more transient and movement quickly returns above baseline. Suppressors and rebounders did not differ on measures of looking during habituation, but when the stimulus changed rebounders looked more than suppressors. When it did not change, they looked less. In addition, during habituation rebounders spent more time looking away from the stimulus. Rapid motor reactivation soon after gaze locks onto a target, characteristic of rebounders, may influence visual foraging and the response to change by keeping attention near a threshold of engagement.     

The Role of Negative Information in Distributional Semantic Learning

Distributional models of semantics learn word meanings from contextual co‐occurrence patterns across a large sample of natural language. Early models, such as LSA and HAL (Landauer & Dumais, 1997; Lund & Burgess, 1996), counted co‐occurrence events; later models, such as BEAGLE (Jones & Mewhort, 2007), replaced counting co‐occurrences with vector accumulation. All of these models learned from positive information only: Words that occur together within a context become related to each other. A recent class of distributional models, referred to as neural embedding models, are based on a prediction process embedded in the functioning of a neural network: Such models predict words that should surround a target word in a given context (e.g., word2vec; Mikolov, Sutskever, Chen, Corrado, & Dean, 2013). An error signal derived from the prediction is used to update each word's representation via backpropagation. However, another key difference in predictive models is their use of negative information in addition to positive information to develop a semantic representation. The models use negative examples to predict words that should not surround a word in a given context. As before, an error signal derived from the prediction prompts an update of the word's representation, a procedure referred to as negative sampling. Standard uses of word2vec recommend a greater or equal ratio of negative to positive sampling. The use of negative information in developing a representation of semantic information is often thought to be intimately associated with word2vec's prediction process. We assess the role of negative information in developing a semantic representation and show that its power does not reflect the use of a prediction mechanism. Finally, we show how negative information can be efficiently integrated into classic count‐based semantic models using parameter‐free analytical transformations.     

The relative salience of discrete and continuous quantity in young infants

Whether human infants spontaneously represent number remains contentious. Clearfield & Mix (1999 ) and Feigenson, Carey & Spelke (2002 ) put forth evidence that when presented with small sets of 1–3 items infants may preferentially attend to continuous properties of stimuli rather than to number, and these results have been interpreted as evidence that infants may not have numerical competence. Here we present three experiments that test the hypothesis that infants prefer to represent continuous variables over number. In Experiment 1, we attempt to replicate the Clearfield & Mix study with a larger sample of infants. Although we replicated their finding that infants attend to changes in contour length, infants in our study attended to number and perimeter/area simultaneously. In Experiments 2 and 3, we pit number against continuous extent for exclusively large sets (Experiment 2) and for small and large sets combined (Experiment 3). In all three experiments, infants noticed the change in number, suggesting that representing discrete quantity is not a last resort for human infants. These results should temper the conclusion that infants find continuous properties more salient than number and instead suggest that number is spontaneously represented by young infants, even when other cues are available.     

Habituation and novelty detection fNIRS brain responses in 5‐ and 8‐month‐old infants: The Gambia and UK

The first 1,000 days of life are a critical window of vulnerability to exposure to socioeconomic and health challenges (i.e. poverty/undernutrition). The Brain Imaging for Global Health (BRIGHT) project has been established to deliver longitudinal measures of brain development from 0 to 24 months in UK and Gambian infants and to assess the impact of early adversity. Here results from the Habituation‐Novelty Detection (HaND) functional near‐infrared spectroscopy (fNIRS) task at 5 and 8 months are presented (N = 62 UK; N = 115 Gambia). In the UK cohort distinct patterns of habituation and recovery of response to novelty are seen, becoming more robust from 5 to 8 months of age. In The Gambia, an attenuated habituation response is evident: a larger number of trials are required before the response sufficiently suppresses relative to the response during the first presented trials. Furthermore, recovery of response to novelty is not evident at 5 or 8 months of age. As this longitudinal study continues in The Gambia, the parallel collection of socioeconomic, caregiving, health and nutrition data will allow us to stratify how individual trajectories of habituation and recovery of response to novelty associate with different risk factors and adaptive mechanisms in greater depth. Given the increasing interest in the use of neuroimaging methods within global neurocognitive developmental studies, this study provides a novel cross‐culturally appropriate paradigm for the study of brain responses associated with attention and learning mechanisms across early development.     

10‐month‐old infants’ inference of invisible agent: Distinction in causality between object motion and human action1

Abstract: Two habituation experiments investigated 10‐month‐old infants’ interpretation of events where a stationary object began to move without any visible causes. During habituation, infants saw that an object partly hidden by an occluder began to move away from the occluder. Then, they were tested with three test events without the occluder: the ?rst event showed a hand pushing the object, the second event showed a hand failing to touch the object, and the last event had no agent. The objects were a ball in Experiment 1, and a person in Experiment 2. The test event that the infants looked at for the shortest duration in Experiment 1 was where the hand pushed the ball, whereas they looked at the three test events almost equal amounts of time in Experiment 2. These results indicate that 10‐month‐old infants responded to the events in terms of causality and could infer the presence of the agent behind the occluder only when they saw the habituation event featuring the ball.     

Investigating looking and social looking measures as an index of infant violation of expectation

Accumulated looking time has been widely used to index violation of expectation (VoE) response in young infants. But there is controversy concerning the validity of this measure, with some interpreting infant looking behaviour in terms of perceptual preferences (Cohen & Marks, 2002 ; Haith, 1998 ). The current study aimed to compare the use of looking time with a recently used measure of social looking (Walden et al., 2007 ) in distinguishing between 6‐month‐old infants’ response to novelty/familiarity and a condition in which the object was covertly switched for a different object. Following habituation, infants showed more social looking in response to the object‐switch condition than the novel object change, whereas the more commonly used accumulated looking time measure did not distinguish between the two, showing an increase for both. Thus, social looking is a more valid measure of infant VoE than looking time.     

Biracial and monoracial infant own‐race face perception: an eye tracking study

We know that early experience plays a crucial role in the development of face processing, but we know little about how infants learn to distinguish faces from different races, especially for non‐Caucasian populations. Moreover, it is unknown whether differential processing of different race faces observed in typically studied monoracial infants extends to biracial infants as well. Thus, we investigated 3‐month‐old Caucasian, Asian and biracial (Caucasian‐Asian) infants’ ability to distinguish Caucasian and Asian faces. Infants completed two within‐subject, infant‐controlled habituation sequences and test trials as an eye tracker recorded looking times and scanning patterns. Examination of individual differences revealed significant positive correlations between own‐race novelty preference and scanning frequency between eye and mouth regions of own‐race habituation stimuli for Caucasian and Asian infants, suggesting that facility in own‐race face discrimination stems from active inspection of internal facial features in these groups. Biracial infants, however, showed the opposite effect: An ‘own‐race’ novelty preference was associated with reduced scanning between eye and mouth regions of ‘own‐race’ habituation stimuli, suggesting that biracial infants use a distinct approach to processing frequently encountered faces. Future directions for investigating face processing development in biracial populations are discussed. A video abstract of this article can be viewed at http://youtu.be/a_dDXfFuEfY     

Neurogenetic approaches to habituation and dishabituation in Drosophila

We review work in the major model systems for habituation in Drosophila melanogaster, encompassing several sensory modalities and behavioral contexts: visual (giant fiber escape response, landing response); chemical (proboscis extension reflex, olfactory jump response, locomotory startle response, odor-induced leg response, experience-dependent courtship modification); electric (shock avoidance); and mechanical (leg resistance reflex, cleaning reflex). Each model system shows several of Thompson and Spencer’s [Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neuronal substrates of behavior. Psychological Review, 73, 16–43] parametric criteria for habituation: spontaneous recovery and dishabituation have been described in almost all of them and dependence of habituation upon stimulus frequency and stimulus intensity in the majority. Stimulus generalization (and conversely, the delineation of stimulus specificity) has given insights into the localization of habituation or the neural architecture underlying sensory processing.The strength of Drosophila for studying habituation is the range of genetic approaches available. Mutations have been used to modify specific neuroanatomical structures, ion channels, elements of synaptic transmission, and second-messenger pathways. rutabaga and dunce, genes of the cAMP signal pathway that have been studied most often in the reviewed experiments, have also been implicated in synaptic plasticity and associative conditioning in Drosophila and other species including mammals. The use of the Gal4/UAS system for targeting gene expression has enabled genetic perturbation of defined sets of neurons. One clear lesson is that a gene may affect habituation differently in different behaviors, depending on the expression, processing, and localization of the gene product in specific circuits. Mutations of specific genes not only provide links between physiology and behavior in the same circuit, but also reveal common mechanisms in different paradigms of behavioral plasticity. The rich repertoire of models for habituation in the fly is an asset for combining a genetic approach with behavioral, anatomical and physiological methods with the promise of a more complete understanding.     

The sustained effect of emotional signals on neural processing in 12‐month‐olds

Around the end of the first year of life, infants develop a social referencing ability – using emotional information from others to guide their own behavior. Much research on social referencing has focused on changes in behavior in response to emotional information. The present study was an investigation of the changes in neural responses that underlie social referencing behavior, reflected in event‐related potentials (ERPs). Twenty‐six 12‐month‐olds participated in a single‐session visit in which ERPs were recorded both immediately before and after a behavioral intervention in which infants' caregivers provided positive, negative or neutral information about each of three stimuli (ERP data available for n = 17). After the intervention, infants devoted more neural resources to processing negative versus neutral and positive information, as observed in early and late positive‐going components. Changes in neural responses from the pre‐ to post‐intervention recordings clarify this observation, indicating a sustained response in the negative and positive conditions, and a decrease in the neutral condition, suggesting an attenuation effect in the neutral condition. Further, infants who attended most to the objects in the behavioral intervention showed increased neural responses in the negative condition and decreased responses in the positive condition. Taken together, these findings suggest that infants' neural responses are differentially affected by positive, negative and neutral information. Furthermore, the findings highlight the importance of measuring the change in neural responses to better interpret post‐experience responses.