Allen Cheung(1), Suzanna Becker (2) and Neil Burgess (1)
1:Department of Anatomy and Institute of Cognitive Neuroscience, UCL
2:Department of Psychology, McMaster University
We present a computational model of the neural mechanisms in the
parietal and temporal lobes that support spatial navigation, imagery,
and episodic recall. Long term representations are stored in the
hippocampus and associated with object and texture information in the
surrounding medial temporal lobe. Translation into a
viewpoint-dependent representation occurs in parietal cortex to enable
construction of an imagined retinotopic scene and generation of
appropriate body movements. This model allows recall and imagery of
locations and objects and selection of the appropriate direction of
movement in complex environments. The model can execute simple
navigational planning strategies by either heading directly for the
goal, or by searching for the next best direction to get around
obstacles. Imagined movement through a familiar environment can occur
in our model by shifting activations on the egocentric parietal map in
a stereotyped way analogous to optic flow, using hard-wired
connections that are gated by the movement signal; if the movement is
a large advance, use of the temporal module is then required to fill
in the locations of new objects that may be brought into view.
Finally, by damaging part of the parietal model, we have simulated
hemispatial neglect in mental imagery that rotates with the imagined
perspective of the observer, as in the famous Milan Square experiment
reported by Bisiach and Luzatti (1978).
Acknowledgements: NB is a Royal Society University Research Fellow.
This work was supported by a research grant from the MRC to NB and
a research grant from NSERC, Canada to SB.