/[gzz]/manuscripts/Paper/paper.tex
ViewVC logotype

Diff of /manuscripts/Paper/paper.tex

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.71 by jvk, Sun Mar 30 17:39:21 2003 UTC revision 1.72 by jvk, Sun Mar 30 19:19:37 2003 UTC
# Line 1268  performance decreases very little over t Line 1268  performance decreases very little over t
1268    
1269  In order to evaluate the recognizability of our procedurally generated  In order to evaluate the recognizability of our procedurally generated
1270  textures, we need to have an appropriate comparison point.  textures, we need to have an appropriate comparison point.
 %  
 %- need to compare to something  
 %- can't generate an infinite supply  
 %  of images easily and there are too much semantics  
1271  Pictures of natural objects would not provide a good comparsion,  Pictures of natural objects would not provide a good comparsion,
1272  because they cannot be generated in infinite amounts from seed  because they cannot be generated in infinite amounts from seed
1273  numbers and they would esasily yield undesirable semantic associations.  numbers and they would esasily yield undesirable semantic associations.
# Line 1279  Thus, we simply use plain solid color ba Line 1275  Thus, we simply use plain solid color ba
1275  even though the colors of even a small set of randomly chosen colors would  even though the colors of even a small set of randomly chosen colors would
1276  most likely not be discriminable.  most likely not be discriminable.
1277    
 %even though know solid colors not really recognizable.  
   
1278  Surely it would be hard to remember a large number of textures, too.  Surely it would be hard to remember a large number of textures, too.
1279  However, the user does not have to remember all the textures;  However, the user does not have to remember all the textures;
1280  it suffices to learn the textures of the most often used documents  it suffices to learn the textures of the most often used documents
# Line 1377  for the textures than for the solid colo Line 1371  for the textures than for the solid colo
1371  %The recognition of solid colors was only almost statistically significantly  %The recognition of solid colors was only almost statistically significantly
1372  %better than random [F(1,4) = 5.7, p = .075].  %better than random [F(1,4) = 5.7, p = .075].
1373  %Whereas ``correct > incorrect'' for textures: [F(1,4) = 24.7, p = .0077]  %Whereas ``correct > incorrect'' for textures: [F(1,4) = 24.7, p = .0077]
1374  The solid colors do not seem to have enough variation for unambiguous  The large number of false positives shows that
1375    solid colors do not have enough variation for unambiguous
1376  recognition.  recognition.
1377  % RT less for correct responses: [F(1,4) = 5.2, p = .082]  % RT less for correct responses: [F(1,4) = 5.2, p = .082]
1378  % RT different for textures and colors: [F(1,4) = .0, p > .1]  % RT different for textures and colors: [F(1,4) = .0, p > .1]
# Line 1388  recognition. Line 1383  recognition.
1383  %also easily be associated with the document content.  %also easily be associated with the document content.
1384  %  %
1385  Of course,  Of course,
1386  the solid colors could facilitate recognition in the presence of other cues,  solid colors could facilitate recognition in the presence of other cues,
1387  such as the text of the document. However, the recognition of  such as the text of the document. However, the recognition of
1388  similar-looking fragments of documents would still depend on conscious effort  similar-looking fragments of documents would still depend on conscious effort
1389  on the user's part ---  on the user's part ---

Legend:
Removed from v.1.71  
changed lines
  Added in v.1.72

savannah-hackers-public@gnu.org
ViewVC Help
Powered by ViewVC 1.1.26