1. Drafts of some reports, in pdf:
"A system-level model of spatial working memory"
- A brief account of the current model.
"The dynamic brain model", in pdf,
- Discussion of computational theoretical issues underlying the model.
"The psychiatric significance of spatial working memory"
- Notes on the research literature on spatial working memory.
"Myths for brain modeling"
- Notes on underlying assumptions.
"Brain modeling and cognitive phenomics"
- How brain modeling might be used for psychiatric categorization
in terms of cognitive measures.
2. The implementation of the model:
A guide to the various directories and files.
The BAD language
- in which the specific model is programmed.
Sicstus Prolog - the underlying programming language.
3. Initial results:
I have done just one run so far, for a basic spatial working
memory experiment. It runs for 306 cycles at 20 milliseconds per cycle,
which is until the hand movement response stops.
(a) input data for the experiment:
(i) the world
(ii) the temporal sequence of stimuli
(b) Trace printing
(i) for the subject
a complete trace,
listing each rule instantiation, the beginning of the run has
the string "init." and each cycle starts with the string "cycle number".
(ii) for the state of the world
on each cycle
(c) VRML graphical visualization sequence showing the state of the brain
model on each cycle as a 3D diagram
also a single
VRML file
for a complete bihemispheric brain model
for cycle 130, showing a more complete 3D visualization.
(d) MRI predicted imaging data
All the activation levels for all modules and for all cycles
are saved, and from this one can generate a file suitable
for input to the Afni MRI imaging system
for any particular cycles one is interested in. It is currently
taking 30 minutes to generate each afni file. As an example,
I generated the afni output file for cycle 130, which is during the delay step
of the experiment.
Sample screen shots, showing the color map, with
Afni images sagittal, axial and coronal view for cycle 130,
I have also generated a
sequence of afni images in axial section
4. Progress report
talk