The Dynamic Field Theory (DFT) is a neural network model of spatial working memory (SWM) that explains changes in spatial memory in real time (Spencer & Schoner, 2000). The proposed research tests a new framework that extends the DFT to spatial language. Preliminary research shows that this process approach can account for contradictory data and shed light on the representational structures underlying language and space. The framework connects with established empirical research in spatial language through reference frames and proposes a dynamic coupling between the SWM field of the DFT and a 2-dimensional label location field. This dynamic coupling predicts that stable or biased activation patterns in one field will lead to similar activation patterns in the other field and thus impact linguistic and non-linguistic behaviors in similar ways. Experiments 1 and 2 test the bounds of this dynamic relationship by investigating connections from the SWM field to 2-D label-location field. Experiments 3 and 4 test these bounds in the other direction from the 2-D label-location field to the SWM field. This empirical work will lay the essential groundwork for development of a formal model linking linguistic and non-linguistic representations of space.
| Lipinski, John; Spencer, John P; Samuelson, Larissa K (2010) Biased feedback in spatial recall yields a violation of delta rule learning. Psychon Bull Rev 17:581-8 |
