PERCEPTION 



'659 



in its boldness and simplicity. The invocation of 

 graded potentials is much more defensible at present 

 than it might have seemed at the time the theory 

 was first proposed. Nevertheless, Lashley (308) and 

 others have stressed a number of difficulties with the 

 theory, the only one so far "that has ever predicted 

 correctly" any perceptual fact. 



First, effects of lesions in the primary projection 

 fields would seem to speak against the theory in its 

 present form. Illusions and constancies are preserved 

 in small remnants of a visual field; motion and pat- 

 terns are perceived across scotomata (see above). 

 Crisscrossing the visual cortex in cats and monkeys 

 with knife cuts, or implanting multiple pieces of gold 

 foil (309, 442) or mica plates (441 ) into the structure, 

 produce no disturbance of pattern vision that is 

 demonstrable with the tests employed. Even the pas- 

 sage of currents into the occiput of human observers 

 who were engaged in various visual tasks (482 ) has 

 thus far produced no measurable change not even 

 in such delicate phenomena as the spontaneous 

 fluctuations of reversible figures. 



The theory, as Kohler himself poinls out, is re- 

 stricted. It is difficult to apply to the tactile-kinesthetic 

 forms of aftereffects (274), or to the visual aftereffects 

 involving displacement of test patterns in (he third 

 dimension (275). [There may be less difficulty in 

 applying the theory to auditory allcrcllects 1 ku, 

 290)]. Tfie greatest restriction of the theory, however, 

 flows from tfie simple fact thai most of the aftereffects 

 described can be obtained with freely moving gaze 

 (Held, unpublished observations). Active scanning 

 of the various patterned fields shown in figure 39 

 induces systematic changes in subsequent perception 

 (97; Held, unpublished observations; see also 410). 

 These observations do not fit tfie present concept of 

 isomorphism [nor, for thai matter, the alternative 

 theory of aftereffects proposed bv Osgood iV 1 lever 

 (371) and based on the earlier work of Marsh, ill & 

 Talbot (340)]. The aftereffects of inspecting espe- 

 cially patterned fields witfi moving gaze seem to be 

 closely connected witfi tfiose set up by prisms or 

 pseudopfiones; a successful theory would eventually 

 have to comprise both sets of phenomena, the figural 

 aftereffects and the adaptation to distorting media. 84 



zi It would be desirable in this connection to determine 

 whether lower species, e.g. fish, show habituation (i.e. decre- 

 ments) to the Muller-Lycr pattern. Fish are subject to these 

 geometric illusions as much as man, and ptrhaps more, as 

 proved by Herter (206). For birds (domestic chicks) there are 

 similar data on strong effects of certain optic-geometric il- 

 lusions (397). For the lemur (an arboreal primate), however, 



FIG. 40. Double Necker cube. On continued fixation at the 

 small horizontal cross-bar in the center, the two cubes fluctuate 

 "spontaneously.' reversing their perspective, often in synchrony, 

 sometimes asynchronously. [From Cohen (90).] 



It would be particularly helpful if we knew whether 

 cortical lesions influence either of these phenomena in 

 some systematic fashion. 86 Unfortunately, information 

 on this point is scant and contradictorv (443). In- 

 creased susceptibility to tactile-kinesthetic afteref- 

 fects in brain-injured men lias been claimed (256 

 and denied (-'371. Bv contrast, reversal rales for 

 ambiguous patterns are known to be influenced by 

 brain injur) (90, 182), but the alterations are difficult 

 to interpret in tin- absence of information on related 

 perceptual phenomena. 



In tfie mosl complete studv of figure reversals 

 after brain injury, reversal rales were found to be 

 decreased after unilater.il lesions in any cerebral 

 lobe [see Cohen (90)], although this decrease was 

 iter with right than with left hemisphere lesion 

 (see fig. 40, 41 ). Bilateral lesions in tfie posterior brain 

 substance diminished reversals even further, Inn 

 bilateral frontal lesions had a paradoxical effect 

 (fig. 411. After this injury, there was marked and 

 persistent enhancement in reversals for the ambiguous 

 pattern, so that bifrontal cases differed from normal 

 controls by showing "too many' reversals, and uni- 

 lateral frontals, bv showing "too few" (fig. 411. Tfie 

 sensitivit) of tliis simple perceptual task to brain 

 injuries of varying sites is remarkable, but tfie inter- 



von Allesch 1494 I reports a "reversal" of the vertical-horizontal 

 illusion, so that this monkey, in contrast to man, overestimates 

 horizontal rather than vertical extents. 



36 Kohler and his associates (2/3, 276) have noted that the 

 transient effects of visual 'satiation' in normal fields tend to 

 resemble some of the persistent changes after occipital lesions 

 in man. In both conditions there is rapid fading of contours, 

 reduced fusion thresholds for flicker and changes in motion 

 perception, and there are displacements and recessions of 

 objects in perceptual space. 



