Early word-learning entails reference, not merely associations

doi:10.1016/j.tics.2009.03.006

Trends in Cognitive Sciences

Volume 13, Issue 6, June 2009, Pages 258-263

https://doi.org/10.1016/j.tics.2009.03.006 Get rights and content

Recent years have witnessed a resurgence of classic tensions concerning the fundamental nature of human knowledge and the processes underlying its acquisition. This tension, especially evident in research on the acquisition of words and concepts, arises when researchers pit one type of content against another (perceptual versus conceptual) and one type of process against another (associative versus theory-based). But these dichotomies are false; they rest upon insufficient consideration of the structure and diversity of the words and concepts that we naturally acquire. As infants and young children establish categories and acquire words to describe them, they take advantage of both perceptual and conceptual information, and relate this to both the (rudimentary) theories they hold and the statistics that they witness.

Section snippets

Two metaphors of development

Two different metaphors undergird recent work on early cognitive and language development. The ‘child-as-data-analyst’ metaphor captures human infants’ impressive capacity to attend to statistical regularities in their environments 1, 2, and the rich sensory, perceptual and computational resources that they bring to the task of acquisition. The ‘child-as-theorist’ metaphor captures infants’ impressive array of conceptual capacities, including core knowledge of physical objects, skeletal

Focusing exclusively on the child-as-data-analyst

In their version of the child-as-data-analyst view, Sloutsky and colleagues 16, 17, 18, 19 have promoted a strict and exclusively associationist approach to early word learning and conceptual development. This work rests on three core assumptions: that the only building blocks for words and concepts are sensory and perceptual experiences, that these experiences are operated upon strictly by general-purpose processes (including associative learning, similarity assessment and attentional

Retaining a place for the child-as-theorist

In our view, capturing the processes underlying early word-learning and conceptual development requires that we also consider the child-as-theorist model. Our concern reflects more than an empirical disagreement; it reflects a fundamentally different set of assumptions concerning words, concepts and development. We draw upon a rich intellectual history within psychology, linguistics and philosophy 20, 21, 22 to focus on four crucial points, which are each considered in turn:

•
Words do not merely

Conclusions

We have underscored that two metaphors – child-as-data-analyst and child-as-theorist – are at play in word-learning and conceptual development. As infants and young children build a repertoire of concepts and acquire words to describe them, they take advantage of both perceptual and conceptual information, and rely upon both the rudimentary theories that they hold and the statistics that they witness. Our goal in writing this article is to emphasize that our theories of acquisition should do

Acknowledgements

This work was supported by NICHD R01HD030410 (S.W.) and NSF BCS-0817128, NICHD R01HD036043 and R56HD036043HD (S.G.). We thank E. Leddon, E. Ware, J. Woodring and four anonymous reviewers for comments on an earlier draft.

References (55)

V.M. Sloutsky
The role of similarity in the development of categorization
Trends Cogn. Sci.
(2003)
A. Cimpian et al.
Preschool children's use of cues to generic meaning
Cognition
(2008)
S.A. Gelman et al.
Young children are sensitive to how an object was created when deciding what to name it
Cognition
(2000)
D. Kelemen
Functions, goals and intentions: children's teleological reasoning about objects
Trends Cogn. Sci.
(1999)
J.E. Opfer et al.
Causal relations drive young children's induction, naming, and categorization
Cognition
(2007)
S.R. Waxman et al.
Words as invitations to form categories: evidence from 12- to 13-month-old infants
Cognit. Psychol.
(1995)
A.L. Fulkerson et al.
Words (but not tones) facilitate object categorization: evidence from 6- and 12-month-olds
Cognition
(2007)
S.R. Waxman et al.
Consistent (but not variable) names as invitations to form object categories: new evidence from 12-month-old infants
Cognition
(2005)
F. Xu et al.
Sensitivity to sampling in Bayesian word learning
Dev. Sci.
(2007)
D.H. Rakison et al.
Developing object concepts in infancy: An associative learning perspective
SRCD Monographs
(2008)

R. Baillargeon

Innate ideas revisited: for a principle of persistence in infants’ physical reasoning

Perspectives on Psychological Science

(2008)

E.S. Spelke

Core knowledge

Am. Psychol.

(2000)

S. Carey

The origin of concepts

(2009)

Gelman, R. and Williams, E.M. (1998). Enabling constraints for cognitive development and learning: domain specificity...

Wellman, H.M. and Gelman, S.A. (1998). Knowledge acquisition. In Handbook of child psychology: cognition, perception,...

Gelman, S.A. and Kalish, C.W. (2006). Conceptual development. In Handbook of child psychology: cognition, perception,...

Waxman, S.R. and Lidz, J.L. (2006). Early word learning. In Handbook of child psychology: cognition, perception, and...

A. Gopnik et al.

Causal learning: psychology, philosophy, and computation

(2007)

Hollich, G.J. et al. (2000). Breaking the language barrier: an emergentist coalition model for the origins of word...

E. Colunga et al.

From the lexicon to expectations about kinds: a role for associative learning

Psychol. Rev.

(2005)

A.E. Booth et al.

Deja vu all over again: re-re-visiting the conceptual status of early word learning: Comment on Smith and Samuelson (2006)

Dev. Psychol.

(2006)

L.K. Samuelson et al.

The shape of controversy: what counts as an explanantion of development? Introduction to the Special Section

Dev. Sci.

(2008)

Smith, L.B. and Samuelson, L. (2006) An attentional learning account of the shape bias: Reply to Cimpian and Markman...

V.M. Sloutsky

When looks are everything: appearance similarity versus kind information in early induction

Psychol. Sci.

(2007)

V.M. Sloutsky

How much does a shared name make things similar? Linguistic Labels, Similarity and the Development of Inductive Inference

Child Dev.

(2001)

V.M. Sloutsky et al.

The role of words and sounds in visual processing: from overshadowing to attentional tuning

Cogn. Sci.

(2008)

A. Gopnik et al.

Words, thoughts, and theories

(1997)

Cited by (207)

Neural evidence for referential understanding of object words in dogs
2024, Current Biology
Using words to refer to objects in the environment is a core feature of the human language faculty. Referential understanding assumes the formation of mental representations of these words.¹^,² Such understanding of object words has not yet been demonstrated as a general capacity in any non-human species,³ despite multiple behavior-based case reports.⁴^,⁵^,⁶^,⁷^,⁸^,⁹^,¹⁰ In human event-related potential (ERP) studies, object word knowledge is typically tested using the semantic violation paradigm, where words are presented either with their referent (match) or another object (mismatch).¹¹^,¹² Such mismatch elicits an N400 effect, a well-established neural correlate of semantic processing.¹²^,¹³ Reports of preverbal infant N400 evoked by semantic violations¹⁴ assert the use of this paradigm to probe mental representations of object words in nonverbal populations. Here, measuring dogs’ (Canis familiaris) ERPs to objects primed with matching or mismatching object words, we found a mismatch effect at a frontal electrode, with a latency (206–606 ms) comparable to the human N400. A greater difference for words that dogs knew better, according to owner reports, further supported a semantic interpretation of this effect. Semantic expectations emerged irrespective of vocabulary size, demonstrating the prevalence of referential understanding in dogs. These results provide the first neural evidence for object word knowledge in a non-human animal.
Neurobiological mechanisms for language, symbols and concepts: Clues from brain-constrained deep neural networks
2023, Progress in Neurobiology
Neural networks are successfully used to imitate and model cognitive processes. However, to provide clues about the neurobiological mechanisms enabling human cognition, these models need to mimic the structure and function of real brains. Brain-constrained networks differ from classic neural networks by implementing brain similarities at different scales, ranging from the micro- and mesoscopic levels of neuronal function, local neuronal links and circuit interaction to large-scale anatomical structure and between-area connectivity. This review shows how brain-constrained neural networks can be applied to study in silico the formation of mechanisms for symbol and concept processing and to work towards neurobiological explanations of specifically human cognitive abilities. These include verbal working memory and learning of large vocabularies of symbols, semantic binding carried by specific areas of cortex, attention focusing and modulation driven by symbol type, and the acquisition of concrete and abstract concepts partly influenced by symbols. Neuronal assembly activity in the networks is analyzed to deliver putative mechanistic correlates of higher cognitive processes and to develop candidate explanations founded in established neurobiological principles.
How “is” shapes “ought” for folk-biological concepts
2022, Cognitive Psychology
Knowing which features are frequent among a biological kind (e.g., that most zebras have stripes) shapes people’s representations of what category members are like (e.g., that typical zebras have stripes) and normative judgments about what they ought to be like (e.g., that zebras should have stripes). In the current work, we ask if people’s inclination to explain why features are frequent is a key mechanism through which what “is” shapes beliefs about what “ought” to be. Across four studies (N = 591), we find that frequent features are often explained by appeal to feature function (e.g., that stripes are for camouflage), that functional explanations in turn shape judgments of typicality, and that functional explanations and typicality both predict normative judgments that category members ought to have functional features. We also identify the causal assumptions that license inferences from feature frequency and function, as well as the nature of the normative inferences that are drawn: by specifying an instrumental goal (e.g., camouflage), functional explanations establish a basis for normative evaluation. These findings shed light on how and why our representations of how the natural world is shape our judgments of how it ought to be.
The scope of infants' early object word extensions
2022, Cognition
Recent research indicates that 6- to 9-month-olds understand a number of object words, but the nature of this understanding is unclear. This work examined whether infants restrict these terms to individual objects (like proper names) or extend them across multiple objects from a category (like common nouns). Experiment 1 reports evidence that 6-month-olds comprehend the name for their mother (e.g., “Mommy”) as restricted to the individual person. Experiment 2 offers support for the claim that 6- and 9-month-olds understand both a label that is restricted to an individual person (e.g., “Mommy”) and a label that extends to multiple members of an object category (i.e., “hand” or “ball”). Experiment 3 provides evidence that 12- to 15-month-olds comprehend both a word that is restricted to an individual (e.g., “Fido”) and a word that extends to multiple category members (e.g., “dog”) for the same object (i.e., a pet dog). The findings indicate that infants understand both individual- and categorical-scope words early in development, suggesting that neither lexical type represents a privileged starting point in word learning. We propose that cross-situational learning abilities, along with intuitive biases to conceptualize objects from particular semantic classes as either individuals or members of categories, play a role in infants' learning of words of the two lexical types.
Event-related potential studies of cross-situational word learning in four-year-old children
2022, Journal of Experimental Child Psychology
Young children learn word-referent links in cross-situational learning paradigms despite uncertainty as to a given word’s correct referent on individual exposures. However, the semantic status of these word-referent links is unknown. Here, we used a novel event-related potential (ERP) testing approach to investigate whether children showed electrophysiological signatures of semantic representations following a typical cross-situational learning paradigm. In Study 1, 20 4-year-old children (mean age: 51.35 months) learned eight new word-object pairs in a cross-situational learning task which included familiar object labeling before training. Then, they completed an ERP task which involved seeing pictures of the novel objects and then hearing words that were either congruent or incongruent with the word-referent association that was trained. Results showed a robust N400 ERP effect, thereby providing evidence for having acquired a semantic representation during the cross-situational learning paradigm. Children also showed good performance on a comprehension test approximately 20 min after training. In a second study, a different sample of 20 4-year-old children (mean age: 52 months) completed the same cross-situational task but without the initial familiar word pretraining. Here, we did not see the N400 effect we saw in Study 1, but children still showed good performance on the delayed comprehension test. We conclude that rapid semantic encoding can occur during cross-situational learning, but that pretraining on the structure of the paradigm may be critical for facilitating that encoding. We speculate on the mechanisms by which pretraining might have effects and discuss the implications of these findings for early word learning.
Mechanisms of associative word learning: Benefits from the visual modality and synchrony of labeled objects
2022, Cortex
Citation Excerpt :
Although it may seem that there is an overlap between the mechanisms behind word learning in infancy and adulthood, there are in fact discrepancies. While word learning in infancy has been argued to merely yield associative links between words and objects (cf. Sloutsky et al., 2017), there is also evidence that there may be mechanisms at work that go beyond association including various conceptually-based computations (Waxman & Gelman, 2009). Adults who learn novel words, with their fully developed cognitive system, may rely on the latter to a larger degree (e.g., hypothesis testing, cf. Xu & Tenenbaum, 2007).
Despite humans' ability to communicate about concepts relating to different senses, word learning research tends to largely focus on labeling visual objects. Although sensory modality is known to influence memory and learning, its specific role for word learning remains largely unclear. We investigated associative word learning in adults, that is the association of an object with its label, by means of event-related brain potentials (ERPs). We evaluated how learning is affected by object modality (auditory vs visual) and temporal synchrony of object-label presentations (sequential vs simultaneous). Across 4 experiments, adults were, in training phases, presented either visual objects (real-world images) or auditory objects (environmental sounds) in temporal synchrony with or followed by novel pseudowords (2 × 2 design). Objects and pseudowords were paired either in a consistent or an inconsistent manner. In subsequent testing phases, the consistent pairs were presented in matching or violated pairings. Here, behavioral and ERP responses should reveal whether consistent object–pseudoword pairs had been successfully associated with one another during training. The visual-object experiments yielded behavioral learning effects and an increased N400 amplitude for violated versus matched pairings indicating short-term retention of object–word associations, in both the simultaneous and sequential presentation conditions. For the auditory-object experiments, only the simultaneous, but not the sequential presentation, revealed similar results. Across all experiments, we found behavioral and ERP correlates of associative word learning to be affected by both sensory modality and partly, by temporal synchrony of object-label combinations. Based on our findings, we argue for independent advantages of temporal synchrony and visual modality in associative word learning.

View all citing articles on Scopus

View full text

Trends in Cognitive Sciences

OpinionEarly word-learning entails reference, not merely associations

Section snippets

Two metaphors of development

Focusing exclusively on the child-as-data-analyst

Retaining a place for the child-as-theorist

Conclusions

Acknowledgements

Trends Cogn. Sci.

Cognition

Cognition

Trends Cogn. Sci.

Cognition

Cognit. Psychol.

Cognition

Cognition

Sensitivity to sampling in Bayesian word learning

Dev. Sci.

Developing object concepts in infancy: An associative learning perspective

SRCD Monographs

Innate ideas revisited: for a principle of persistence in infants’ physical reasoning

Perspectives on Psychological Science

Core knowledge

Am. Psychol.

The origin of concepts

Causal learning: psychology, philosophy, and computation

From the lexicon to expectations about kinds: a role for associative learning

Psychol. Rev.

Deja vu all over again: re-re-visiting the conceptual status of early word learning: Comment on Smith and Samuelson (2006)

Dev. Psychol.

The shape of controversy: what counts as an explanantion of development? Introduction to the Special Section

Dev. Sci.

When looks are everything: appearance similarity versus kind information in early induction

Psychol. Sci.

How much does a shared name make things similar? Linguistic Labels, Similarity and the Development of Inductive Inference

Child Dev.

The role of words and sounds in visual processing: from overshadowing to attentional tuning

Cogn. Sci.

Words, thoughts, and theories

Opinion
Early word-learning entails reference, not merely associations