| 1 |
Advancing Our Understanding of Early Perceptual and Cognitive Development Essay Review of Emerging Cognitive Abilities in Early Infancy
|
|
|
|
In: http://www.homepages.ucl.ac.uk/~ucjtml0/docs/bertenthal%26longo-humdev-2002.pdf
|
|
BASE
|
|
Show details
|
|
|
| 2 |
Productivity, proceduralisation and SLI: Comment on Hsu and Bishop
|
|
|
|
In: http://nrl.northumbria.ac.uk/4630/1/productivityproceduralisationandsli.pdf
|
|
Abstract:
Abstract Hsu and Bishop argue that the underlying deficit in SLI is impaired statistical/procedural learning (in particular, difficulties in learning non-adjacent dependencies). This prevents affected children from developing abstract syntactic representations which in turn leads to diminished productivity. However, grammatical productivity need not be based on abstract syntactic representations. Moreover, abstraction is a matter of degree, and most everyday language use relies on relatively concrete, low-level schemas. I suggest that the grammatical problems in SLI may be due not to failure to develop abstract representations, but to a disruption of later consolidation processes which support the development of fast, automatic and errorfree performance. Productivity, proceduralisation and SLI: Comment on Hsu and Bishop In the 1990s, many researchers tried to explain SLI by appealing to an innately specified "language module", which was assumed to be impaired in affected individuals. Over the years, language researchers have become increasingly dissatisfied with this explanation, and number of alternative accounts have been put forward which attempt to explain the language difficulties in SLA as a consequence of an underlying processing or learning deficit. Hsu and Bishop's contribution is part of this research tradition. Specifically, they argue that "… grammatical deficits arise when the learning system is biased towards the memorisation of exemplars, and is poor at extracting statistical dependencies from the input…. SLI involves deficits in extracting nonadjacent dependencies from input, leading to reliance on rote learning." (abstract) This is an interesting proposal, and Hsu and Bishop cite an impressive amount of suggestive evidence. However, while the general idea is promising, the view of language and language learning that they assume is based on some questionable assumptions. I begin this commentary by discussing these implicit assumptions, and then show how their basic insight can be reformulated in a usage-based theory of language acquisition. Grammatical productivity needn't be based on abstract representations Hsu and Bishop draw a sharp distinction between concrete (exemplar-based) representations on the one hand and "system-wide abstract syntactic schemes" (5) on the other, and argue that the latter are needed "to support production and accurate comprehension of sentences that a child never heard before" (5). However, we do not need to postulate abstract syntactic structures in order to explain linguistic productivity: novel forms can also be constructed by using a stored exemplar as a model, and applying analogy (KISS+PAST is to kissed as HISS+PAST is to ???). Exemplar models work particularly well for morphology (e.g. Eddington, 2000), but have also been applied successfully to a number of syntactic phenomena (see In usage-based approaches, schemas are regarded as abstract symbolic units which capture the relational similarities between the concrete exemplars from which they emerge. Since analogy also relies on relational similarity, applying analogy and extracting schemas are closely linked. Applying analogy involves three stages: retrieving relevant exemplars from memory (which requires assessing the similarity of the target to potential models), establishing correspondences between matching parts, and computing the novel form. If a speaker repeatedly retrieves the same, or substantially overlapping, set of exemplars for the purposes of computing an analogy and compares them to establish correspondences between subparts, the links between the corresponding subparts will be strengthened, resulting in the emergence of a schema. Thus, schema extraction can be regarded as a result of applying analogy to produce and understand novel forms (
|
|
URL: http://nrl.northumbria.ac.uk/4630/1/productivityproceduralisationandsli.pdf
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1083.5942
|
|
BASE
|
|
Hide details
|
|
|
| 3 |
Unruptured Cerebral Aneurysm Detected after Intravenous Tissue Plasminogen Activator
|
|
|
|
In: ftp://ftp.ncbi.nlm.nih.gov/pub/pmc/54/c1/Case_Rep_Neurol_2009_Jun_20_1(1)_20-23.tar.gz
|
|
BASE
|
|
Show details
|
|
|
| 4 |
A Randomized Controlled Trial on Very Early Speech and Language Therapy in Acute Stroke Patients with Aphasia
|
|
|
|
In: ftp://ftp.ncbi.nlm.nih.gov/pub/pmc/bb/5a/Cerebrovasc_Dis_Extra_2011_Jul_12_1(1)_66-74.tar.gz
|
|
BASE
|
|
Show details
|
|
|
| 5 |
Analysis of English Nonsense Syllable Recognition in Noise
|
|
|
|
In: http://www-personal.umich.edu/%7Ebenki/pubs/BenkiPhonetica2003.pdf
|
|
BASE
|
|
Show details
|
|
|
| 6 |
Talk, Small Stories, and Adolescent Identities
|
|
|
|
In: http://www.clarku.edu/~mbamberg/Papers/HD-Reply2004.pdf
|
|
BASE
|
|
Show details
|
|
|
|
|
