An associative account of the development of word learning

Word learning is a notoriously difficult induction problem because meaning is underdetermined by positive examples. How do children solve this problem? Some have argued that word learning is achieved by means of inference: young word learners rely on a number of assumptions that reduce the overall hypothesis space by favoring some meanings over others. However, these approaches have difficulty explaining how words are learned from conversations or text, without pointing or explicit instruction. In this research, we propose an associative mechanism that can account for such learning. In a series of experiments, 4-year-olds and adults were presented with sets of words that included a single nonsense word (e.g. dax). Some lists were taxonomic (i.,e., all items were members of a given category), some were associative (i.e., all items were associates of a given category, but not members), and some were mixed. Participants were asked to indicate whether the nonsense word was an animal or an artifact. Adults exhibited evidence of learning when lists consisted of either associatively or taxonomically related items. In contrast, children exhibited evidence of word learning only when lists consisted of associatively related items. These results present challenges to several extant models of word learning, and a new model based on the distinction between syntagmatic and paradigmatic associations is proposed.     

The interrelatedness between infants’ communicative gestures and lexicon size: A longitudinal study

Research has shown a close relationship between gestures and language development. In this study, we investigate the cross-lagged relationships between different types of gestures and two lexicon dimensions: number of words produced and comprehended. Information about gestures and lexical development was collected from 48 typically developing infants when these were aged 0;9, 1;0 and 1;3. The European Portuguese version of the MacArthur–Bates Communicative Development Inventory: Words and Gestures (PT CDI:WG) was used. The results indicated that the total number of actions and gestures and the number of early gestures produced at 0;9 and at 1;0 year predicted the number of words comprehended three months later. Actions and gestures’ predictive power of the number of words produced was limited to the 0;9–1;0 year interval. The opposite relationship was not found: word comprehension and production did not predict action and gestures three months later. These results highlight the importance of non-verbal communicative behavior in language development.     

The role of phonological recoding in orthographic learning

The objective of this research was to explore whether orthographic learning occurs as a result of phonological recoding, as expected from the self-teaching hypothesis. The participants were 32 fourth- and fifth-graders (mean age = 10 years 0 months, SD = 7 months) who performed lexical decisions for monosyllabic real words and pseudowords under two matched experimental conditions: a read aloud condition, wherein items were named prior to lexical decision to promote phonological recoding, and a concurrent articulation condition, presumed to attenuate phonological recoding. Later, orthographic learning of the pseudowords was evaluated using orthographic choice, spelling, and naming tasks. Consistent with the self-teaching hypothesis, targets learned with phonological recoding in the read aloud condition yielded greater orthographic learning than those learned with concurrent articulation. The research confirms the critical nature of phonological recoding in the development of visual word recognition skills and an orthographic lexicon.     

Simulating single word processing in the classic aphasia syndromes based on the Wernicke-Lichtheim-Geschwind theory

The Wernicke-Lichtheim-Geschwind (WLG) theory of the neurobiological basis of language is of great historical importance, and it continues to exert a substantial influence on most contemporary theories of language in spite of its widely recognized limitations. Here, we suggest that neurobiologically grounded computational models based on the WLG theory can provide a deeper understanding of which of its features are plausible and where the theory fails. As a first step in this direction, we created a model of the interconnected left and right neocortical areas that are most relevant to the WLG theory, and used it to study visual-confrontation naming, auditory repetition, and auditory comprehension performance. No specific functionality is assigned a priori to model cortical regions, other than that implicitly present due to their locations in the cortical network and a higher learning rate in left hemisphere regions. Following learning, the model successfully simulates confrontation naming and word repetition, and acquires a unique internal representation in parietal regions for each named object. Simulated lesions to the language-dominant cortical regions produce patterns of single word processing impairment reminiscent of those postulated historically in the classic aphasia syndromes. These results indicate that WLG theory, instantiated as a simple interconnected network of model neocortical regions familiar to any neuropsychologist/neurologist, captures several fundamental "low-level" aspects of neurobiological word processing and their impairment in aphasia.     

Sound-symbolism: A Piece in the Puzzle of Word Learning

Sound-symbolism is the idea that the relationship between word sounds and word meaning is not arbitrary for all words, but rather that there are subsets of words in the world’s languages for which sounds and their symbols have some degree of correspondence. The present research investigates sound-symbolism as a possible route to the learning of an unknown word’s meaning. Three studies compared the guesses that adult participants made regarding the potential meanings of sound-symbolic and non-sound symbolic obsolete words. In each study, participants were able to generate better definitions for sound-symbolic words when compared to non-sound symbolic words. Participants were also more likely to recognize the meanings of sound symbolic words. The superior performance on sound-symbolic words held even when definitions generated on the basis of sound association were eliminated. It is concluded that sound symbolism is a word property that influences word learning.     

Novel labels support 10-month-olds’ attention to novel objects

What is the source of the mutual exclusivity bias whereby infants map novel labels onto novel objects? In an intermodal preferential looking task, we found that novel labels support 10-month-olds’ attention to a novel object over a familiar object. In contrast, familiar labels and a neutral phrase gradually reduced attention to a novel object. Markman (1989, 1990) argued that infants must recall the name of a familiar object to exclude it as the referent of a novel label. We argue that 10-month-olds’ attention is guided by the novelty of objects and labels rather than knowledge of the names for familiar objects. Mutual exclusivity, as a language-specific bias, might emerge from a more general constraint on attention and learning.     

Phonological recoding,rapid automatized naming,and orthographic knowledge

Phonological recoding, orthographic knowledge, and rapid automatized naming (RAN) are three major contributors to word identification. However, the interrelations between these components remain somewhat unclear. The current analyses focus on how phonological recoding and alphanumeric versus non-alphanumeric RAN contribute to different components of orthographic knowledge (word specific vs. general). Results indicate that alphanumeric and non-alphanumeric RAN contribute to orthographic knowledge components differently. Alphanumeric RAN relates more to word-specific orthographic knowledge, whereas non-alphanumeric RAN relates more to general orthographic knowledge. Furthermore, phonological recoding is more closely related to word-specific orthographic knowledge than to general orthographic knowledge.     

Detecting Orientation of a Surface: The Rectangularity Postulate and Primary Depth Cues

Computational metaphors for determining the orientation of planar surfaces represented in line drawings have exploited a postulate that often surfaces are rectangular. Previous research implies that people follow such logic with real surfaces in ecological viewing. However, this research is problematic methodologically and some research does not directly address the issue. The stimuli used in this study were rectangular and trapezoidal; the latter shape was used to mislead with regard to orientation under the rectangularity postulate. Viewing conditions were monocular and binocular, with and without observer movement. The results suggest that the rectangularity postulate was important under stationary monocular viewing but diminished with movement and was not apparent during binocular viewing. General arguments about the importance of secondary depth cues in ecological viewing are developed.     

The neural correlates of statistical learning in a word segmentation task: An fMRI study

Functional magnetic resonance imaging (fMRI) was used to assess neural activation as participants learned to segment continuous streams of speech containing syllable sequences varying in their transitional probabilities. Speech streams were presented in four runs, each followed by a behavioral test to measure the extent of learning over time. Behavioral performance indicated that participants could discriminate statistically coherent sequences (words) from less coherent sequences (partwords). Individual rates of learning, defined as the difference in ratings for words and partwords, were used as predictors of neural activation to ask which brain areas showed activity associated with these measures. Results showed significant activity in the pars opercularis and pars triangularis regions of the left inferior frontal gyrus (LIFG). The relationship between these findings and prior work on the neural basis of statistical learning is discussed, and parallels to the frontal/subcortical network involved in other forms of implicit sequence learning are considered.     

The temporal dynamics of first versus second language production

In this study we explored the temporal origin of processing differences between first and second language production. Forty highly proficient bilinguals named objects of high and low lexical frequency aloud for both L1 and L2 separately while event-related brain potentials (ERPs) were recorded. The first electrophysiological differences elicited by response language occurred at the same early P2 peak (∼140–220 ms) where we observed the onset of the lexical frequency effect, but only for those bilinguals who started naming in an L1 context and afterwards switched to an L2 naming context. The bilinguals who named objects in the reverse direction did not display a language effect in the ERPs. Taken together, the data show that the L2 naming disadvantage originates during the onset of lexical access and seems to be driven by both representational strength, which is lower for L2 words, and language control demands, which are higher for L2 words.