Language - Computational Explorations in Cognitive Neuroscience - page 333

Visual
Vis + Sem
Semantic
Blend
Other

Visual
Vis + Sem
Semantic
Blend
Other

Abstract

Concrete Concrete

Abstract

SP_Hid

SP_Hid 20 No Sem No Sem

20

15 15

10

10

5

5

0

0

OS_Hid OS_Hid

20 20 Full Sem Full Sem

15

15

10

10

5

5

0 0

0.0 0.2 0.4 0.6 0.8

Lesion Proportion

0.0 0.2 0.4 0.6 0.8

Lesion Proportion

0.0 0.2 0.4 0.6 0.8

Lesion Proportion

0.0 0.2 0.4 0.6 0.8

Lesion Proportion

Figure 10.12: Error types for direct pathway lesions, with
both an intact semantic pathway (Full Sem) , and with a com-
plete semantic pathway lesion (No Sem).

Figure 10.11: Error types for semantic pathway lesions in
conjunction with a completely lesioned direct pathway. N =

for each point.

more semantic errors on abstract words relative to con-
crete words.

Figure 10.11 shows the same semantic pathway le-
sions as the previous figure, in conjunction with a com-
plete lesion of the direct pathway (i.e., lesion types 3
and 4). This corresponds to the type of lesion studied
by Plaut and Shallice (1993) in their model of deep
dyslexia. For all levels of semantic pathway lesion, we
now see semantic errors, together with visual errors and
a relatively large number of “other” (uncategorizable)
errors. This pattern of errors is generally consistent with
that of deep dyslexia, where all of these kinds of errors
are observed. Comparing figure 10.11 with the previ-
ous figure, we see that the direct pathway was playing
an important role in generating correct responses, par-
ticularly in overcoming the semantic confusions that the
semantic pathway would have otherwise made.

Question 10.6 (a) Is there evidence in the model for a
difference between concrete and abstract words in the
number of semantic errors made? (b) Explain why this
occurs in terms of the nature of the semantic represen-
tations in the model for these two types of words (recall
that concrete words have richer semantics with more
overall units).

Figure 10.12 shows the effects of direct pathway le-
sions, both with (lesion type 2) and without (lesion type
5) an intact semantic pathway. Let's focus first on the
case with the intact semantic pathway (the Full Sem
graphs in the figure). Notice that for smaller levels of
damage (i.e., about 40% and below), relatively few or
no semantic errors are produced, with most of the errors
being visual. This pattern corresponds well with phono-
logical dyslexia, especially assuming that this damage
to the direct pathway interferes with the pronunciation
of nonwords, which can presumably only be read via
this direct orthography to phonology pathway. Unfor-
tunately, we can't test this aspect of the model because
the small number of training words provides an insuffi-
cient sampling of the regularities that underlie success-
ful nonword generalization, but the large-scale model
of the direct pathway described in the next section pro-

Question 10.5 Compare the first bar in each graph of
figure 10.11 (corresponding to the case with only a di-
rect pathway lesion, and no damage to the semantic
pathway) with the subsequent bars. (a) Does addi-
tional semantic pathway damage appear to be neces-
sary to produce the semantic error symptoms of deep
dyslexia? (b) Explain why the direct pathway lesion
leads to semantic errors.

Figure 10.11 also shows the relative number of se-
mantic errors for the concrete versus abstract words.
One characteristic of deep dyslexia is that patients make