Computational Linguistics & Phonetics Computational Linguistics & Phonetics Fachrichtung 4.7 Universität des Saarlandes

Distinguished Speakers in Language Science

Thursday, 15 January 2015, 12:15
Conference Room, Building C7.4

Unusual time!

Nonlinear dynamics of lifelong learning: The myth of cognitive decline

Michael Ramscar
Department of Linguistics
University of Tübingen

As adults age, their reaction times slow across a range of psychometric tests. This has been widely taken to show that cognitive information–processing capacities decline over the course of adulthood. I will show that these response patterns, which are typically taken as evidence for (and measures of) declining cognitive–processing capacities, arise naturally out of basic principles of learning. These basic, formal learning principles both correctly identify the pattern of performance exhibited by very young children in word learning tasks (which can differ greatly from that of young adults) as well successfully predicting that older adults will exhibit far greater sensitivity to fine–grained properties of test stimuli than younger adults. Taken together, the findings I present show that the patterns of change observed in cognitive performance across the lifespan simply reflect the consequences of learning from the kind of statistical distributions that typify human experience. Once the information–processing loads are inevitably imposed by learning from this experience are controlled for, it appears that the performance changes that are usual taken as evidence of innate abilities in infants, or declining cognitive capacities in older adults, support little more than the unsurprising idea that the way individuals choose between or recalls items will be affected by the number of items there are and what an individual has already learned about them. I will consider the implications of this for our scientific and cultural understanding of lifelong cognitive development.

If you would like to meet with the speaker, please contact Vera Demberg.