Chris Sinha

Symbol Grounding or the Emergence of Symbols? Vocabulary Growth in Children and a Connectionist Net

Abstract The Symbolic Grounding Problem is viewed as a by-product of the classical cognitivist approach to studying the mind. In contrast, an epigenetic interpretation of connectionist approaches to studying the mind is shown to offer an account of symbolic skills as an emergent, developmental phenomenon. We describe a connectionist model of concept formation and vocabulary growth that auto-associates image representations and their associated labels. The image representations consist of clusters of random dot figures, generated by distorting prototypes. Any given label is associated with a cluster of random dot figures. The network model is tested on its ability to reproduce image representations given input labels alone (comprehension) and to identify labels given input images alone (production). The model implements several well-documented findings in the literature on early semantic development; the occurrence of over- and under-extension errors; a vocabulary spurt; a comprehension/production asymmetr...

Infant search tasks reveal early concepts of containment and canonical usage of objects.

Abstract It is difficult to gain unambiguous evidence on the use of concepts by infants. Many results can be accounted for in terms of action-based strategies. The evidence reported here fulfils the minimal criteria for the operation of working concepts in infants. Search tasks are used with a filled interval which forces memory-search, and the object is hidden in containers which fulfil their customary job or violate it. Infants treat an upright cup as a more reliable location marker than an inverted one. A series of experiments probes the phenomenon. The results indicate that the infants have a working concept of containment which can be triggered by the provision of containers in their canonical orientation. Even “object permanence tasks” lead infants to access their knowledge of the relationships into which things typically enter in the world outside the laboratory.

Distributed Spatial Semantics

The “local semantics” approach to the analysis of the meaning of locative particles (e.g. spatial prepositions) is examined, criticized and rejected. An alternative, distributed approach to spatial relational semantics and its linguistic expression is argued for. In the first part of the paper, it is argued that spatial relational semantic information is not exclusively carried in languages such as English by the locative particle, and that “item-specific meanings plus selectional restrictions” cannot save the localist approach. In the second part of the paper, the “covertly” distributed spatial relational semantics of languages such as English is contrasted with the “overtly” distributed spatial relational semantics characterizing many other languages. Some common assumptions relating to the universality of the expression of spatial relational meaning by closed syntactic classes are criticized. A change of perspective from “local” to “distributed” semantics permits the re-analysis of polysemy and item-bound “use-type” in terms of the distributed expression of language-specific spatial relational semantic types.

When time is not space: the social and linguistic construction of time intervals and temporal event relations in an Amazonian culture

Abstract It is widely assumed that there is a natural, prelinguistic conceptual domain of time whose linguistic organization is universally structured via metaphoric mapping from the lexicon and grammar of space and motion. We challenge this assumption on the basis of our research on the Amondawa (Tupi Kawahib) language and culture of Amazonia. Using both observational data and structured field linguistic tasks, we show that linguistic space-time mapping at the constructional level is not a feature of the Amondawa language, and is not employed by Amondawa speakers (when speaking Amondawa). Amondawa does not recruit its extensive inventory of terms and constructions for spatial motion and location to express temporal relations. Amondawa also lacks a numerically based calendric system. To account for these data, and in opposition to a Universal Space-Time Mapping Hypothesis, we propose a Mediated Mapping Hypothesis, which accords causal importance to the numerical and artefact-based construction of time-based (as opposed to event-based) time interval systems.

Comparative Spatial Semantics and Language Acquisition: Evidence from Danish, English, and Japanese

Spatial relational meaning is typically predominantly expressed in English and related languages by the locative particle system. Even between closely related languages such as Danish and English, there are substantial differences with respect to both the semantics and the morphology of locative particles. Other languages (including Japanese), although they may use locative particles in spatial relational expression, distribute spatial relational meaning quite differently between and within form classes. We investigate the consequences of these differences for the acquisition of spatial relational expressions in these three languages. Although the structure of the target language affects the specific strategies employed by the language acquiring child, the patterns of production development in all three languages appear to be instances of a general class of conservative learning strategies. We discuss the implications of these findings in terms of the relationship between linguistic and cognitive determinants of spatial language acquisition

Learning through interaction: Language as interaction

‘Not to let a word get in the way of its sentence Nor to let a sentence get in the way of its intention, But to send your mind out to meet the intention as a guest; THAT is understanding.’ Chinese proverb, fourth century b.c. Most people, if asked what a language is, would almost certainly answer in terms of ‘sounds’, ‘words’ and ‘sentences’. They would probably also refer to something less clearly defined which they might call ‘meaning’. And they might just possibly add something about the purposes that language – both spoken and written – serves in the interpersonal transactions that constitute so large a part of everyday life. Such an ordering of priorities no doubt owes much to the way in which ‘language’ is encountered during the process of education: in dictionaries, in the form of comprehension exercises, and in lessons on grammar and spelling. It also corresponds quite closely to the relative emphasis that has been given to the various aspects of language in the long tradition of serious study that goes back as far as Aristotle and even earlier. The same emphasis on sounds, words and sentences, treated as units within a formal system, has also characterised the greater part of the work carried out in the present century by linguists and others who have attempted to study language ‘scientifically’.

Spatial reference systems and the canonicality effect in infant search

Abstract Infants have a bias in search tasks toward reaching centrally in their visual field rather than peripherally. When care is taken to control this, another bias has been found: one toward treating upright cups as more reliable location markers for a hidden object than inverted cups. This has been named the “canonicality effect”. In this study, the two biases are elicited in the same sample of infants. When a transposition design resulted in the hidden object maintaining either a central or a constant peripheral position, infants searched more reliably centrally than peripherally, regardless of the orientation of the cups hiding the object. However, this bias toward centrality did not hold when the position of the hidden object was not held constant in relation to its initial position: when transposition resulted in the object shifting from a central to a peripheral position, or vice versa, the infants shifted search more reliably with upright cups than with inverted ones regardless of the central-peripheral variable. The two biases seem to be alternatives. Infants apparently switch between spatial codes, rather than employing both codes across all conditions. It was found possible to reverse the canonicality effect under predicted conditions in a way which confirms its analysis as a local spatial encoding of the point of search, that is as a relation between object and hiding place.

Time, space, and events in language and cognition: a comparative view

We propose an event‐based account of the cognitive and linguistic representation of time and temporal relations. Human beings differ from nonhuman animals in entertaining and communicating elaborate detached (as opposed to cued) event representations and temporal relational schemas. We distinguish deictically based (D‐time) from sequentially based (S‐time) representations, identifying these with the philosophical categories of A‐series and B‐series time. On the basis of cross‐linguistic data, we claim that all cultures employ both D‐time and S‐time representations. We outline a cognitive model of event structure, emphasizing that this does not entail an explicit, separate representation of a time dimension. We propose that the notion of an event‐independent, metric “time as such” is not universal, but a cultural and historical construction based on cognitive technologies for measuring time intervals. We critically examine claims that time is universally conceptualized in terms of spatial metaphors, and hypothesize that systematic space–time metaphor is only found in languages and cultures that have constructed the notion of time as a separate dimension. We emphasize the importance of distinguishing what is universal from what is variable in cultural and linguistic representations of time, and speculate on the general implications of an event‐based understanding of time.

The allative bias in three-year-olds is almost proof against task naturalness

An allative bias has been noted in childrens comprehension of deictic verbs and directional prepositions. This study investigates the allative bias and its susceptibility to contextual modification. Three-year-olds were asked to act according to requests to move one object to or from another, and to identify where an object had come from or gone to. A reliable bias was found whereby ‘to’ was easier than ‘from’. Attempts to make the utterance more naturally related to the task, by using movement and fronted objects, modulated the bias to a small extent, but did not abolish it. An analysis of directional expressions in terms of the components SOURCE, PATH and GOAL is presented in relation to the task-demands.

Language and other artifacts: socio-cultural dynamics of niche construction.

Niche construction theory is a relatively new approach in evolutionary biology that seeks to integrate an ecological dimension into the Darwinian theory of evolution by natural selection. It is regarded by many evolutionary biologists as providing a significant revision of the Neo-Darwinian modern synthesis that unified Darwin’s theory of natural and sexual selection with 20th century population genetics. Niche construction theory has been invoked as a processual mediator of social cognitive evolution and of the emergence and evolution of language. I argue that language itself can be considered as a biocultural niche and evolutionary artifact. I provide both a general analysis of the cognitive and semiotic status of artifacts, and a formal analysis of language as a social and semiotic institution, based upon a distinction between the fundamental semiotic relations of “counting as” and “standing for.” I explore the consequences for theories of language and language learning of viewing language as a biocultural niche. I suggest that not only do niches mediate organism-organism interactions, but also that organisms mediate niche-niche interactions in ways that affect evolutionary processes, with the evolution of human infancy and childhood as a key example. I argue that language as a social and semiotic system is not only grounded in embodied engagements with the material and social-interactional world, but also grounds a sub-class of artifacts of particular significance in the cultural history of human cognition. Symbolic cognitive artifacts materially and semiotically mediate human cognition, and are not merely informational repositories, but co-agentively constitutive of culturally and historically emergent cognitive domains. I provide examples of the constitutive cognitive role of symbolic cognitive artifacts drawn from my research with my colleagues on cultural and linguistic conceptualizations of time, and their cultural variability. I conclude by reflecting on the philosophical and social implications of understanding artifacts co-agentively.