AG, a pure case of optic anemia (object naming impaired; action naming
good) is described. We consider the fit of experimental data from AG
to different theoretical accounts of optic aphasia. Overall, we find n
o evidence for impairments intrinsic to semantic representations, but
we note a number of problems that we interpret as indicating a slight,
and specific, weakness in semantic access from vision. We also note a
mild problem in generating names to a cue (verbal fluency). The main
aim of the report was to provide a full description of tests of visual
, semantic, and speech output skills in this patient in relation to th
e processing of visually presented objects and scenes, within a sequen
tial information processing account which enables some theoretical imp
lications to be drawn, albeit not conclusively. (C) 1996 Academic Pres
s, Inc.