Categorization in infancy: labeling induces a persisting focus on commonalities

Althaus, Nadja; Plunkett, Kim

doi:10.1111/desc.12358

Cited by 33 publications

(38 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, infants successfully learn object categories from a mixture of labeled and unlabeled exemplars; although it is unclear from this prior work whether infants integrated the unlabeled exemplars or simply ignored them (Balaban & Waxman, ; Waxman & Markow, ). Second, substantial evidence suggests the effect of naming extends beyond the exemplars that have been named: naming directs infants' attention to commonalities among objects and, moreover, influences their interpretation of as‐yet‐unnamed objects (Althaus & Mareschal, ; Althaus & Plunkett, ; Waxman & Braun, ; Waxman & Markow, ). Perhaps, then, exposure to labeled exemplars provides infants with a robust foundation for learning from subsequent, unlabeled exemplars.…”

Section: Introductionmentioning

confidence: 99%

A little labeling goes a long way: Semi‐supervised learning in infancy

LaTourrette

Waxman

2018

Developmental Science

View full text Add to dashboard Cite

There is considerable evidence that labeling supports infants' object categorization. Yet in daily life, most of the category exemplars that infants encounter will remain unlabeled. Inspired by recent evidence from machine learning, we propose that infants successfully exploit this sparsely labeled input through "semi-supervised learning." Providing only a few labeled exemplars leads infants to initiate the process of categorization, after which they can integrate all subsequent exemplars, labeled or unlabeled, into their evolving category representations. Using a classic novelty preference task, we introduced 2-year-old infants (n = 96) to a novel object category, varying whether and when its exemplars were labeled. Infants were equally successful whether all exemplars were labeled (fully supervised condition) or only the first two exemplars were labeled (semi-supervised condition), but they failed when no exemplars were labeled (unsupervised condition). Furthermore, the timing of the labeling mattered: when the labeled exemplars were provided at the end, rather than the beginning, of familiarization (reversed semi-supervised condition), infants failed to learn the category. This provides the first evidence of semi-supervised learning in infancy, revealing that infants excel at learning from exactly the kind of input that they typically receive in acquiring real-world categories and their names.

show abstract

Section: Introductionmentioning

confidence: 99%

A little labeling goes a long way: Semi‐supervised learning in infancy

LaTourrette

Waxman

2018

Developmental Science

View full text Add to dashboard Cite

show abstract

“…Equally, labels heard in‐the‐moment also begin to exert a powerful influence on processing during the first year. For example, the in‐task presence of a novel label can direct ten‐month‐old infants’ attention to commonalities between category exemplars and guide online category formation (Althaus & Plunkett, 2015; Plunkett, Hu, & Cohen, 2008), and labels themselves facilitate the formation of new representations over other auditory cues (e.g., Althaus & Westermann, 2016; for a review, see Robinson, Best, Deng, & Sloutsky, 2012). …”

mentioning

confidence: 99%

Learned Labels Shape Pre‐speech Infants’ Object Representations

Twomey

Westermann

2017

Infancy

View full text Add to dashboard Cite

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.

show abstract

“…Behavioral and looking times studies (e.g., Althaus & Plunkett, 2016) report a minimum of six trials as habituation criterion whereas our randomization did not allow us to reach this criterion. This was done to prevent infants from habituating to a particular action sequence or condition.…”

Section: Methodsmentioning

confidence: 99%

Syntactical regularities of action sequences in the infant brain: when structure matters

Maffongelli

Antognini

Daum

2018

Developmental Science

View full text Add to dashboard Cite

Throughout life, actions and language are inherent to social interactions. A long-standing research question in cognitive neuroscience concerns the interrelation between verbal and non-verbal forms of social interactions, that is, language and action. Perceiving how actions are performed and why they are performed in a certain way is crucial for the observer to anticipate the actor's goal and to prepare an appropriate response. It is suggested that predicting upcoming events in a given action sequence can be compared to the way we process the language information flow. Goal-directed actions can be sequenced in small units, which are organized according to a hierarchical plan, resembling the hierarchical organization of language. Research on adults suggests that manipulating the action structure (i.e., action syntax) leads to analogous cortical signatures as a similar manipulation of a sentence structure (i.e., language syntax). Whereas in adults language and action knowledge are based on life-time experience, in infants both domains are still developing. The current study examined the neural processing of structural violations of observed goal-directed action sequences in infants at 6-7 months, using event-related potentials (ERPs). Results showed that a structural violation of the action sequence elicited bilateral frontal positivity effects. This suggests that infants capture structural regularities, and it adds a crucial element to the understanding of general syntactic regularities and their violation from an ontogenetic perspective.

show abstract

Categorization in infancy: labeling induces a persisting focus on commonalities

Cited by 33 publications

References 19 publications

A little labeling goes a long way: Semi‐supervised learning in infancy

A little labeling goes a long way: Semi‐supervised learning in infancy

Learned Labels Shape Pre‐speech Infants’ Object Representations

Syntactical regularities of action sequences in the infant brain: when structure matters

Contact Info

Product

Resources

About