McClelland Rogers Semantic Cognition 2009
<p>Nov 25, 2009 9:15 PM
 
Semantic Cognition:  Its Nature, Its Development, and Its Neural Basis
 
McClelland, Tim Rogers, Karalyn Patterson,  Karia Dilkina, Matt Ralph
 
Book Chapter 72     2009
from Gazzaniga  The Cognit Neurosciences  IV
 
 
*
 
page 8  Modeling SEMANTIC DEMENTIA - second column
 
    The cortical semantic network adheres to the CONVERGENCE PRINCIPLE
SEMANTIC KNOW of all kinds of concepts, across all modalities of reception and expression depends on
a relativ. circumscibed region of the ANTERIOR TEMP LOBES.
Perhaps the func. role is that of the REPRES LAYER in Rumelhart's network.
 
We believe the difft surface repres. are subserved by difft brain areas organized by sense modality and situated near the sense channels where they receive input.  The func of the semantic system is to mediate learned assoc. among the
various difft surface repres.  So that  when a
 
page 9  line drawing of a pencil is observed:  respres of assoc attributes comes up:  color yellow, anme pencil, action writing
will be activated in the approp modality specific brain area.
 
Our framework differs from others in that we believe that the assocs are mediated by a CENTRAL HUB in
the ATL.
 
page 14:  A SINGLE SYSTEM for KNOW of Objects and WORDS
Clearly SD pts  show a conunction of sematnic defects and lexical defects.
   In semantic tasks, pts lose know of specific and idiosyncratic properties of objects but retain
and overextend general and typical property knowledge.
Similarly with words, they love specific words, meanings, mappings but retian and
overextend the typical.
 
 
<b>page 15  Brain areas involved:  Role of the ATL  (anter tempor lobe)</b>
Their hypothesis is that  ATL lobes  jointly provide  an AMODAL HUB for semantic knowledge.
 
The ATLS provide a hub over which conceputal knowledge is represented.
 
In Senile Dementia,  atrophy of ATL and inferolateeral temp lobes is paired with 
semantic impairment.
 
Somre corroboration in Alzheimers and herpes simp encephalitis  
BUT NO CORROBOARATION from fMRI studies of semantic memory.
 
However,  rTMS  repetitive transcran mag stim  DOES show an effect...
slower recog of on the specific level   golden retriever.
 
Both  R and L  ATL  are = affected by rTMS for semantic mem tasks.
 
However,  patients with ATL RESECTION for epilepsy are not affected wot the same degree as SD pts.
 
<b>17  Analogy:  </b>Using a pocket knife to make a sandwich:  opening a package, slicing bread, meat, cheese,
scooping and spreading mayonnaise  Different aspects of knife must be brought to the fore. 
 
 Must distinguish  semantic REPRES from the CONTROL processes that USE semantic knowledge.
eg  frontal  lobs fts may have semantic aphasia and have diffic with use of knowledge.
Seman Dementia pts have deter in the knowledge itself  whereas  Semantic aphasia pts have
defic in the REGULATION of the know.
 
Studies in nl participants consistently implicate <b>PREFRONTAL CORTEX</b> and<b> temporoparietal junction</b>
in tassk requiring control of semantic processing ... eg when a partic aspect of meaning
must be selected or when there is strong competition from alternative repsonses.
PFC and TPJ are involved in general executive/attential functions even when they do not involve
semantic info.
 
 
 
 
 
 
 
 
 
 
 </p>