[Amdnl] Jean Piaget Society 2012 summary

Tue Jun 19 16:05:16 EDT 2012

Tom, that's fantastic!  I'm particularly excited to see the connectionist
and dynamic systems people together.  :)  It's too bad ICIS and JPS were so
close together in time -- I would have loved to see the symposia.

Angelo is encouraging each of the AMD task forces to think about organizing
a special issue of TAMD.  When you have some time, let's brainstorm a bit
and see if we can propose a topic/theme that will help continue the
momentum.  An idea suggested by Giorgio Metta that would work well is to
solicit papers from pairs of robotics/behavioral researchers who have
collaborated, to not only provide developmental people with examples of
their own paradigms, but also demonstrate what robotics/modeling offers as
a tool kit.

Cheers,

Matt

On Tue, Jun 19, 2012 at 10:43 AM, Thomas Shultz, Dr. <
thomas.shultz at mcgill.ca> wrote:

>  Dear Colleagues:****
>
> Thanks to Matt Schlesinger for his summary of the developmental robotics
> workshop (organized with Jochen Triesch) that took place before the
> International Conference on Infant Studies in Minneapolis: a nice example
> of improving collaboration between computational and behavioral researchers
> interested in development. ****
>
> In a similar vein, I wanted to mention that there were two relevant
> invited symposia at the 42nd Annual Meeting of the Jean Piaget Society,
> held in Toronto 31 May – 2 June. These symposia were well attended and
> enthusiastically received, with considerable audience engagement and
> interaction. It is clear that researchers in cognitive development are not
> only interested in following relevant computational work but also in some
> cases getting involved with it, either as practioners or empirical
> collaborators. ****
>
> Also, Linda Smith gave a relevant and well-attended invited address at
> this meeting. These three JPS events are summarized below.****
>
> Best wishes,****
>
> Tom****
>
> -----------------------------------------------------------------
> Thomas Shultz, Professor, Department of Psychology
> McGill University, 1205 Penfield Ave., Montreal, Quebec,
> Canada H3A 1B1.
> Associate Member, School of Computer Science
> E-mail: thomas.shultz at mcgill.ca****
>
> Updated 25 March 2011:
> http://www.psych.mcgill.ca/perpg/fac/shultz/personal/default.htm
> Phone: 514 398-6139
> Fax: 514 398-4896
> -----------------------------------------------------------------****
>
> ** **
>
> *Invited Symposium 5*
>
> *Saturday 2 June, 3:00-4:30, Essex*
>
> John Spencer <john-spencer at uiowa.edu>****
>
> *Invited Symposium 5: From complex insights to complex systems:
> Rethinking how development happens*
>
> Organizer: John P Spencer (University of Iowa)****
>
> ** **
>
> This symposium brings together several of the top researchers who use
> formal, complex systems approaches to understand how cognition changes over
> development. The goal is to rethink Piaget’s insights in the context of
> modern, formal theories. The central question is whether these modern tools
> give us new leverage to understand the mechanisms that underlie changes in
> thinking over time. The first talk will re-visit associative systems which
> have often been discounted as too simple to capture developmental change.
> The message is that association is a powerful mechanism for creating
> change, even in relatively abstract cognitive tasks. The second talk will
> examine the emergence of new cognitive abilities in early development,
> focusing on the critical question of how working memory capacity increases
> over time. The final talk will provide an overview of constructive
> artificial neural networks that have shed new light on a host of changes in
> cognitive development. Together, the talks will highlight cutting-edge work
> using neural process models to understand how development happens,
> integrating a formal, complex systems perspective with a rigorous approach
> to studying behavioral development. ****
>
> ** **
>
> *Throwing out Lucienne and Laurent with the bathwater: How the
> development of abstraction via perception and action can emerge from
> associative systems** *
>
> Bob McMurray (University of Iowa)****
>
> Keith Apfelbaum (University of Iowa)****
>
> Larissa Samuelson (University of Iowa)****
>
> *Capacity in context: Dynamic processes of behavior, memory, and
> development*
>
>  Vanessa Simmering (University of Wisconsin—Madison)****
>
> *Constructive artificial neural-network models for cognitive development**
> *
>
> Thomas Shultz (McGill University)****
>
> ** **
>
> *Invited Symposium 4*
>
> *Saturday 2 June, 9:00-10:30, Essex*
>
> Thomas Shultz <thomas.shultz at mcgill.ca>****
>
> *Invited Symposium 4: Computational approaches to development*
>
> Organizer: Thomas Shultz (McGill University)****
>
> Computational modeling implements developmental theory in a precise
> manner, allowing generation, explanation, integration, and prediction of
> phenomena. The three most prominent modeling approaches to psychological
> development (neural networks, dynamic systems, and Bayesian) are each well
> represented in this symposium. ****
>
> The classical approach to analogy emphasizes structured representations
> and structure mapping. Instead, Denis Mareschal sees analogy as deriving
> from relational priming. He implements this in a neural network in which
> relations are represented as transformations between states. Network
> performance covers several developmental phenomena, illustrating how a
> complex cognitive skill emerges from priming. ****
>
> Dynamic systems consist of a set of quantitative variables that change
> continually, concurrently, and interdependently over time, corresponding to
> differential equations. John Spencer and Sammy Perone probe the development
> of working and long-term memory in the first year. Their simulations show
> that key aspects of developmental change can be captured by Hebbian
> learning in a system that autonomously explores the visual world. This
> suggests that new cognitive abilities emerge in the first year as infants
> construct their own development via exploratory activity.****
>
> Research on epistemic trust typically interprets children's success as
> evidence that they track others' knowledge. Patrick Shafto presents
> evidence that 4-year-olds' behavior is better explained by reasoning about
> both knowledge and intent, and developmental differences between 3 and 4
> are consistent with changing abilities to reason about intent, not
> knowledge.****
>
> *A connectionist model of analogical completion*
>
> Denis Mareschal (Birkbeck, University of London)****
>
> *Constructing development in an autonomous dynamical system: Insights
> from dynamic field theory*
>
> John P Spencer (University of Iowa)****
>
> Sammy Perone (University of Iowa)****
>
> *Epistemic trust: Modeling children's reasoning about others' knowledge
> and intent*
>
> Patrick Shafto (University of Louisville)****
>
> 3:00-4:15 PL04 Plenary Session
> 4............................................................................................................Osgoode
> East****
>
> ** **
>
> *The sensory-motor dynamics of early word learning*****
>
> Linda Smith (Indiana University)****
>
> The visual world is cluttered with many targets and many distractions. Yet
> very young children readily learn object names from the co-occurrences of
> hearing spoken names with such cluttered scenes. Mapping an as-yet
> unlearned name to the right referent requires selecting and stabilizing
> attention on attention on the intended referent in the scene. This talk
> will present new evidence on how young learners solve these problems
> through an embodied attentional system that involves hands, heads, and
> eyes. More specifically, the talk will present evidence from dynamically
> dense and multimodal measures that show: (1) how toddlers through their own
> hand, head and eye movements select and stabilize visual attention on a
> selected target; (2) how toddlers’ sensorimotor activities significantly
> reduce visual clutter and competition in the visual field, and in so doing
> support word learning; and (3) how parents and toddler form a complex
> system of dynamically coupled hand, head, and eye movements so as to
> achieve tightly coordinated visual attention.****
>
> ** **
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.cse.msu.edu/pipermail/amdnl/attachments/20120619/542210e5/attachment-0001.htm 