Exp. Brain Res. 25, 139-156 (1976)
Experimental Brain Research 9 by Springer-VerlagI976
Recovery of Function in Cat Visual Cortex Following Prolonged Deprivation* M. Cynader 1, N. Berman 2 and A. Hein 3 1 Department of Psychology, Dalhousie University, Halifax, Nova Scotia (Canada) z Department of Anatomy, Washington University, School of Medicine, St. Louis, MO. 63110 (USA) 3 Department of Psychology, Massachusetts Institute of Technology, Cambridge, Mass. 02139 (USA)
Summary. Evidence that there is a critical period during which response characteristics of neurons in visual cortex of the cat may be influenced has been provided in several studies, which suggest that the period of influence is restricted to the first few months of life. Using a somewhat different experimental procedure, we have obtained evidence that cortical units retain plasticity long after the end of this period has passed. In our procedure prolonged visual deprivation was followed by exposure in a normal visual environment. The animals were maintained throughout the first year of life either in total darkness or in an enclosure illuminated intermittently by a strobe light. Following the period of deprivation, electrophysiologic recordings were taken from some of these animals. The remaining cats were permitted 6-12 months in a normally-illuminated environment prior to recording. Cats of the same age reared from birth in a normally lit environment were also recorded. Cortical neurons in cats deprived of any normal visual experience rarely show orientation selective responses. In animals allowed subsequent normal visual experience about one-half of the units studied exhibited this property. This level of response specificity is intermediate between that of normally-reared and recently-deprived animals. While most cortical units in normally-reared cats exhibit direction selectivity, this property is rarely observed in the "recovery" cats. A number of unit types which are rarely observed in either normal or totally deprived animals were encountered in cats that had normal exposure following prolonged deprivation. A convergent strabismus was observed, in contrast with the divergent strabismus often shown by cats immediately following prolonged visual deprivation. This shows that ocular alignment as well as cortical unit properties can remain plastic in the adult. Key words: Environmental modification - Critical period - Visual cortex Electrophysiology. * Supported by NRC Grant No. A9939 and M.R.C. Grant No. MA5201 (to M.C.) and Grants from NIH and the Sloan Foundation (to A.H.).
140
M. Cynaderet al.
Introduction
Exposure in a visually atypical environment during the neonatal period appears to modify response characteristics of single neurons in the visual system of the cat. Rearing kittens in special visual environments has altered the distribution of ocular dominance (Hubel and Wiesel, 1970; Wiesel and Hubel, 1963, 1965), preferred orientations (Blakemore and Cooper, 1970; Hirsch and Spinelli, 1970) and preferred directions (Cynader, Berman and Hein, 1975; Tretter, Cynader and Singer 1975) of cortical neurons. It has also been possible to induce response characteristics which do not occur in normally-reared animals by selective exposure to special visual environments (Cynader, Berman and Hein, 1973; Pettigrew and Freeman, 1973; Van Sluyters and Blakemore, 1973). Much of the recent evidence on environmental modifiability has been reviewed by Barlow (1975). Several studies have delimited a "critical period" before or after which manipulation of the visual environment does not alter the response characteristics of the cortical neurons. Sensitivity to environmental conditions appears to extend from about 3 weeks to 3 months of age in the cat with the maximum effect of brief exposure appearing in the fourth week of life. It has been suggested that in kittens 28 days of age, exposures as short as a few hours can alter the response of cortical neurons (Blakemore and Mitchell, 1974; Tretter, Cynader and Singer, 1975). Once the animal is three months old, the response characteristics of cortical units become relatively immutable (Blakemore and Van Sluyters, 1974; Hubel and Wiesel, 1970) although exposure to a monotonous visual environment combined with visual deprivation appear to induce some alterations in adult animals as well (Creutzfeldt and Heggelund, 1975). Lack of modifiability in receptive-field characteristics of cortical neurons after the critical period contrasts sharply with demonstrations of plasticity at the behavioral level. For example, cats subjected to monocular deprivation throughout the "critical period" show deficits in a variety of tasks when first using the deprived eye. Performance improves when the previously-deprived eye is exposed and the originally-exposed eye is sutured shut (Chow and Stewart, 1972; Dews and Wiesel, 1970). Similarly, cats reared under stroboscopic illumination at low flash frequencies show impaired visuomotor coordination (Hein, Gower and Diamond, 1971). This behavioral result seems consistent with the electrophysiological evidence indicating a cortex with many non-oriented units (Cynader, Berman and Hein, 1973). However, after a period of exposure in a normal visual environment, the visuomotor coordination of these animals improves (Hein et al., 1971). In another case, cats deprived of normal vision for long periods of time by binocular eyelid suture or dark rearing improve in a variety of behavioral tasks after subsequent exposure in a normal environment (Baxter, 1966; Chow and Stewart, 1972; Sherman, 1973; Van Hof-Van Duin, 1975). The relation of single unit characteristics to visual capacities and visuallycoordinated behaviors is, of course, as yet unclear. However, the persistence of a capacity for behavioral change beyond the early months of life might be
Function in Cat Visual Cortex
141
c a u s e d b y c h a n g e s in r e c e p t i v e - f i e l d p r o p e r t i e s o f c o r t i c a l n e u r o n s . T h i s s u g gests that alterations of unit response characteristics during the neonatal period b y e x p o s u r e in a t y p i c a l e n v i r o n m e n t s m a y n o t b e p e r m a n e n t . T o e x p l o r e t h i s possibility we recorded from strobe-reared and light-deprived cats immediately after prolonged deprivation and after a subsequent period of exposure in normal illumination. Data from normally-reared cats are also presented for purposes of comparison.
Methods Subjects The subjects were eleven cats reared from birth until they were 11 to 15 months old under the conditions described. Eight cats were light deprived; 5 by being kept in total darkness and 3 by suture of both eyelids at 3-5 days of age. At the end of the deprivation period, four of the light-deprived animals were studied electrophysiologically (LD group) while the other 4 were permitted exposure in a normally illuminated colony for the next 6-12 months (LD-N group) and then recorded. Three other cats were reared for 1 year in an environment which precluded movement of contours across the retinae. These animals were kept in an enclosure illuminated solely by a brief (10 microsecond) strobe flash once every 2 sec. Subsequently, they were permitted exposure in a normal environment for 6-12 months following the end of the deprivation period (S-N group) and then recorded. Nineteen normally-reared cats (N group) were also recorded.
Procedures The procedures for preparing the animals for unit studies and for visual stimulation, recording, and data reduction and analysis were similar to those used in our other investigations (Berman and Cynader, 1972; Cynader and Berman, 1972). Cats were anaesthetised initially with intravenous sodium pentothal, then intubated by mouth and paralysed with Flaxedil. During recording Pentothal was discontinued and animals were maintained on a mixture of 70% N20 and 30% 02. End-Tidal CO2 and rectal temperature were monitored and maintained at physiologic levels. Following retinoscopic examination appropriate contact lenses and artificial pupils ensured that stimuli presented on a tangent screen 1.5 m distant were in focus on the retinae. Clean conditions were maintained and, at the end of the recording session, the animals were allowed to recover from paralysis and returned to their home cages. Some animals in the S-N and LD-N group were studied up to four times under these conditions, without apparent deterioration of receptive-field properties from session to session.
Orientation and Direction Selectivity While the concepts of direction selectivity and orientation selectivity seem intuitively obvious, it is possible for ambiguities to arise in the categorization of units unless care is taken to distinguish between these properties. We called a unit direction selective if it responded vigorously to stimuli moving in some directions and poorly or not at all to others. We adopted the requirement that stimuli moving in the presumptive preferred direction evoke at least twice as many spikes/presentation as stimuli moving in the opposite direction. In their initial studies, Hubel and Wiesel (1963) considered a unit orientation selective if it responded optimally to a moving slit of a given orientation and not at all to an orthogonally-oriented slit. This criterion, however, would require that many direction-selective units in the cat superior collieulus which respond well to small stimuli of many different shapes and which do not distinguish among tongues with different leading edge configurations (concave, convex, irregular, ser-
142
M. Cynader et al.
rated) be classified as orientation selective. We classified a direction selective unit as orientation selective if it met one of the following specifications: a) it responded well to elongated moving slits, bars, or edges but poorly to small moving spots or irregularly-shaped objects b) its response to flashed slits confined entirely within the activating region of the receptive field varied with slit orientation c) its direction selectivity was considerably narrower for elongated slits or edges than for spots d) it exhibited preferential responses to two directions of movement, 180 ~ apart, tested either with spots or slits, with weaker responses to other directions. Some units exhibited orientation selectivity to flashed stimuli, and responded equally when the optimally oriented slit was moved in two directions 180 ~ apart. These units were classified as orientation selective but not direction selective.
Criteria for Unit Types
Our criteria for unit types are similar to those employed by other workers (Hubel and Wiesel, 1962; Pettigrew, Nikara and Bishop, 1967; Blakemore, Fiorentini and Maffei, 1972). Simple cells were characterised by their small receptive fields (0.5 ~ to 2~ near the area centralis), low spontaneous activity (0-2 spikes/sec), tight velocity tuning (0.5 to 5~ and, when responses to flashed stimuli were present, separable "on" or "off" areas. When tested with moving edges, simple cells had separable leading and trailing edge discharge centers. Units giving only "on" or "off" responses to flashed stimuli but satisfying the other criteria listed above were also classed as simple cells. Complex cells had larger receptive fields (at least 1~ and broader velocity tuning than simple cells. They tended to have higher spontaneous activity and exhibited a much more sustained response as stimuli moved through the receptive field than did simple cells. When responses to flashed stimuli were present, intermingled "on" and "off" areas were observed. The optimum stimulus could be much narrower than the width of the activating region of the receptive field. Hypercomplex cells were characterised by their requirement for a restricted length of stimulus. They responded optimally to stimuli limited in length either on one side of the receptive field (single-stopped) or on both sides (double-stopped). A variety of subtypes of hypercomplex cells have been described (Dreher, 1972; Hubel and Wiesel, 1965). We called units hypercomplex only if short stimuli evoked at least twice as many spikes per presentation as did elongated stimuli. A final category of cells failed to meet the criteria for orientation selectivity but responded preferentially to movement in one direction. Direction selectivity tuning was usually quite broad (halfamplitude width 90 ~ or more). These units were called "pure direction selective" (Blakemore et aL, 1972). Transitional unit types with characteristics of more than one category were observed and other units seemed not to fall into any of the designated groups. These units were termed "unclassified" and are reported separately.
Results I. N o r m a l R e a r i n g R e s p o n s e c h a r a c t e r i s t i c s o f v i s u a l c o r t e x u n i t s in n o r m a l l y - r e a r e d cats h a v e b e e n d e s c r i b e d in d e t a i l b y o t h e r i n v e s t i g a t o r s ( B i s h o p , C o o m b s a n d H e n r y , 1971; Hubel and Wiesel, 1962; Singer, Cynader and Tretter, 1975). Our data a r e in g e n e r a l a g r e e m e n t w i t h t h e s e r e s u l t s a n d p r o v i d e a n a p p r o p r i a t e c o m parison group for our experimental animals. Table 1 summarizes our results f o r t h e n o r m a l l y - r e a r e d cat. O u r o b s e r v a t i o n s o n c o m p l e x cells in n o r m a l l y - r e a r e d cats a r e h o w e v e r r e l e v a n t to t h o s e o f t h e " r e c o v e r y " g r o u p a n d a r e p r e s e n t e d h e r e in m o r e d e t a i l . W e h a v e s u b d i v i d e d c o m p l e x cells i n t o t w o t y p e s d e p e n d i n g o n t h e s t r e n g t h o f
Function in Cat Visual Cortex
143
Table 1. Illustrates the distribution of orientation and direction selectivity among the various clas-
ses of cortical neurons in normally-reared cats and in light-deprived cats (LD groups). The total number of cells is given at the end of the caption and the values in the table are percentages. The percentage of units in each subgroup for the light-deprived cats is given in parenthesis. Normal cat (N = 594), LD cat (N = 103) Oriented units
Directional units
Oriented Oriented non-directional directional Simple 4 (0) Complex 4 (2) Oriented unclassified 3 (0) Hypercomplex 2 (0) Pure direction selective Non-oriented stopped Non-oriented concentric Non-oriented general Totaloriented
3l (6) 24 (4) 13 (0) 2 (0)
83 ( 1 2 )
Non-oriented units
Non-oriented directional
Non-oriented non-directional
9 (1) 3 (1) 1 (7) 2 (17) 20 (62) Totalnon-oriented 17 (88)
Total directional 82 (I 1) their summation along the appropriately oriented line. One type of complex cell responded only to moving slits, and small moving spots and irregularlyshaped objects were ineffective stimuli. The majority of the complex cells encountered however, gave vigorous responses to small spots and other irregularly-shaped objects and responses to these stimuli were often as strong as to the best oriented slit which we could find. The response of a direction-selective complex cell of this second type are displayed in Fig. 1. The left hand side of the figure illustrates the responses of the unit to slits of various orientations. As can be seen, there is a vigorous response to the vertical slit, but the horizontal slit evokes no response as it crosses the receptive field. The responses to small spots moving leftward however, are just as vigorous as those to the appropriately oriented slit. Moreover, the range of directions of movement over which the unit responds is much greater when spots are used. Clear responses can be observed to upward and downward movement of the small spot. We have observed vigorous responses to short slits of the "worst" orientation, provided that the slits are "stopped" in length and completely confined to the activating region of the receptive field. These units thus behave like hypercomplex ceils when confronted with stimuli of inappropriate orientations. At the best orientation, increasing the slit length does not result in more vigorous responses, but there is also no inhibition of the response to elongated contours. The responses to the best oriented slit and to the spot are therefore equal. At the other orientations, the units respond only to short stimuli. Transition units between the summating and non-summating type of complex cell can also be observed. In these units one observes an increment in response as the best oriented slit is made progressively longer, but responses to small spots are often quite vigorous.
144
M. Cynader et al.
Slits
,
Experimental Brain Research 9 by Springer-VerlagI976
Recovery of Function in Cat Visual Cortex Following Prolonged Deprivation* M. Cynader 1, N. Berman 2 and A. Hein 3 1 Department of Psychology, Dalhousie University, Halifax, Nova Scotia (Canada) z Department of Anatomy, Washington University, School of Medicine, St. Louis, MO. 63110 (USA) 3 Department of Psychology, Massachusetts Institute of Technology, Cambridge, Mass. 02139 (USA)
Summary. Evidence that there is a critical period during which response characteristics of neurons in visual cortex of the cat may be influenced has been provided in several studies, which suggest that the period of influence is restricted to the first few months of life. Using a somewhat different experimental procedure, we have obtained evidence that cortical units retain plasticity long after the end of this period has passed. In our procedure prolonged visual deprivation was followed by exposure in a normal visual environment. The animals were maintained throughout the first year of life either in total darkness or in an enclosure illuminated intermittently by a strobe light. Following the period of deprivation, electrophysiologic recordings were taken from some of these animals. The remaining cats were permitted 6-12 months in a normally-illuminated environment prior to recording. Cats of the same age reared from birth in a normally lit environment were also recorded. Cortical neurons in cats deprived of any normal visual experience rarely show orientation selective responses. In animals allowed subsequent normal visual experience about one-half of the units studied exhibited this property. This level of response specificity is intermediate between that of normally-reared and recently-deprived animals. While most cortical units in normally-reared cats exhibit direction selectivity, this property is rarely observed in the "recovery" cats. A number of unit types which are rarely observed in either normal or totally deprived animals were encountered in cats that had normal exposure following prolonged deprivation. A convergent strabismus was observed, in contrast with the divergent strabismus often shown by cats immediately following prolonged visual deprivation. This shows that ocular alignment as well as cortical unit properties can remain plastic in the adult. Key words: Environmental modification - Critical period - Visual cortex Electrophysiology. * Supported by NRC Grant No. A9939 and M.R.C. Grant No. MA5201 (to M.C.) and Grants from NIH and the Sloan Foundation (to A.H.).
140
M. Cynaderet al.
Introduction
Exposure in a visually atypical environment during the neonatal period appears to modify response characteristics of single neurons in the visual system of the cat. Rearing kittens in special visual environments has altered the distribution of ocular dominance (Hubel and Wiesel, 1970; Wiesel and Hubel, 1963, 1965), preferred orientations (Blakemore and Cooper, 1970; Hirsch and Spinelli, 1970) and preferred directions (Cynader, Berman and Hein, 1975; Tretter, Cynader and Singer 1975) of cortical neurons. It has also been possible to induce response characteristics which do not occur in normally-reared animals by selective exposure to special visual environments (Cynader, Berman and Hein, 1973; Pettigrew and Freeman, 1973; Van Sluyters and Blakemore, 1973). Much of the recent evidence on environmental modifiability has been reviewed by Barlow (1975). Several studies have delimited a "critical period" before or after which manipulation of the visual environment does not alter the response characteristics of the cortical neurons. Sensitivity to environmental conditions appears to extend from about 3 weeks to 3 months of age in the cat with the maximum effect of brief exposure appearing in the fourth week of life. It has been suggested that in kittens 28 days of age, exposures as short as a few hours can alter the response of cortical neurons (Blakemore and Mitchell, 1974; Tretter, Cynader and Singer, 1975). Once the animal is three months old, the response characteristics of cortical units become relatively immutable (Blakemore and Van Sluyters, 1974; Hubel and Wiesel, 1970) although exposure to a monotonous visual environment combined with visual deprivation appear to induce some alterations in adult animals as well (Creutzfeldt and Heggelund, 1975). Lack of modifiability in receptive-field characteristics of cortical neurons after the critical period contrasts sharply with demonstrations of plasticity at the behavioral level. For example, cats subjected to monocular deprivation throughout the "critical period" show deficits in a variety of tasks when first using the deprived eye. Performance improves when the previously-deprived eye is exposed and the originally-exposed eye is sutured shut (Chow and Stewart, 1972; Dews and Wiesel, 1970). Similarly, cats reared under stroboscopic illumination at low flash frequencies show impaired visuomotor coordination (Hein, Gower and Diamond, 1971). This behavioral result seems consistent with the electrophysiological evidence indicating a cortex with many non-oriented units (Cynader, Berman and Hein, 1973). However, after a period of exposure in a normal visual environment, the visuomotor coordination of these animals improves (Hein et al., 1971). In another case, cats deprived of normal vision for long periods of time by binocular eyelid suture or dark rearing improve in a variety of behavioral tasks after subsequent exposure in a normal environment (Baxter, 1966; Chow and Stewart, 1972; Sherman, 1973; Van Hof-Van Duin, 1975). The relation of single unit characteristics to visual capacities and visuallycoordinated behaviors is, of course, as yet unclear. However, the persistence of a capacity for behavioral change beyond the early months of life might be
Function in Cat Visual Cortex
141
c a u s e d b y c h a n g e s in r e c e p t i v e - f i e l d p r o p e r t i e s o f c o r t i c a l n e u r o n s . T h i s s u g gests that alterations of unit response characteristics during the neonatal period b y e x p o s u r e in a t y p i c a l e n v i r o n m e n t s m a y n o t b e p e r m a n e n t . T o e x p l o r e t h i s possibility we recorded from strobe-reared and light-deprived cats immediately after prolonged deprivation and after a subsequent period of exposure in normal illumination. Data from normally-reared cats are also presented for purposes of comparison.
Methods Subjects The subjects were eleven cats reared from birth until they were 11 to 15 months old under the conditions described. Eight cats were light deprived; 5 by being kept in total darkness and 3 by suture of both eyelids at 3-5 days of age. At the end of the deprivation period, four of the light-deprived animals were studied electrophysiologically (LD group) while the other 4 were permitted exposure in a normally illuminated colony for the next 6-12 months (LD-N group) and then recorded. Three other cats were reared for 1 year in an environment which precluded movement of contours across the retinae. These animals were kept in an enclosure illuminated solely by a brief (10 microsecond) strobe flash once every 2 sec. Subsequently, they were permitted exposure in a normal environment for 6-12 months following the end of the deprivation period (S-N group) and then recorded. Nineteen normally-reared cats (N group) were also recorded.
Procedures The procedures for preparing the animals for unit studies and for visual stimulation, recording, and data reduction and analysis were similar to those used in our other investigations (Berman and Cynader, 1972; Cynader and Berman, 1972). Cats were anaesthetised initially with intravenous sodium pentothal, then intubated by mouth and paralysed with Flaxedil. During recording Pentothal was discontinued and animals were maintained on a mixture of 70% N20 and 30% 02. End-Tidal CO2 and rectal temperature were monitored and maintained at physiologic levels. Following retinoscopic examination appropriate contact lenses and artificial pupils ensured that stimuli presented on a tangent screen 1.5 m distant were in focus on the retinae. Clean conditions were maintained and, at the end of the recording session, the animals were allowed to recover from paralysis and returned to their home cages. Some animals in the S-N and LD-N group were studied up to four times under these conditions, without apparent deterioration of receptive-field properties from session to session.
Orientation and Direction Selectivity While the concepts of direction selectivity and orientation selectivity seem intuitively obvious, it is possible for ambiguities to arise in the categorization of units unless care is taken to distinguish between these properties. We called a unit direction selective if it responded vigorously to stimuli moving in some directions and poorly or not at all to others. We adopted the requirement that stimuli moving in the presumptive preferred direction evoke at least twice as many spikes/presentation as stimuli moving in the opposite direction. In their initial studies, Hubel and Wiesel (1963) considered a unit orientation selective if it responded optimally to a moving slit of a given orientation and not at all to an orthogonally-oriented slit. This criterion, however, would require that many direction-selective units in the cat superior collieulus which respond well to small stimuli of many different shapes and which do not distinguish among tongues with different leading edge configurations (concave, convex, irregular, ser-
142
M. Cynader et al.
rated) be classified as orientation selective. We classified a direction selective unit as orientation selective if it met one of the following specifications: a) it responded well to elongated moving slits, bars, or edges but poorly to small moving spots or irregularly-shaped objects b) its response to flashed slits confined entirely within the activating region of the receptive field varied with slit orientation c) its direction selectivity was considerably narrower for elongated slits or edges than for spots d) it exhibited preferential responses to two directions of movement, 180 ~ apart, tested either with spots or slits, with weaker responses to other directions. Some units exhibited orientation selectivity to flashed stimuli, and responded equally when the optimally oriented slit was moved in two directions 180 ~ apart. These units were classified as orientation selective but not direction selective.
Criteria for Unit Types
Our criteria for unit types are similar to those employed by other workers (Hubel and Wiesel, 1962; Pettigrew, Nikara and Bishop, 1967; Blakemore, Fiorentini and Maffei, 1972). Simple cells were characterised by their small receptive fields (0.5 ~ to 2~ near the area centralis), low spontaneous activity (0-2 spikes/sec), tight velocity tuning (0.5 to 5~ and, when responses to flashed stimuli were present, separable "on" or "off" areas. When tested with moving edges, simple cells had separable leading and trailing edge discharge centers. Units giving only "on" or "off" responses to flashed stimuli but satisfying the other criteria listed above were also classed as simple cells. Complex cells had larger receptive fields (at least 1~ and broader velocity tuning than simple cells. They tended to have higher spontaneous activity and exhibited a much more sustained response as stimuli moved through the receptive field than did simple cells. When responses to flashed stimuli were present, intermingled "on" and "off" areas were observed. The optimum stimulus could be much narrower than the width of the activating region of the receptive field. Hypercomplex cells were characterised by their requirement for a restricted length of stimulus. They responded optimally to stimuli limited in length either on one side of the receptive field (single-stopped) or on both sides (double-stopped). A variety of subtypes of hypercomplex cells have been described (Dreher, 1972; Hubel and Wiesel, 1965). We called units hypercomplex only if short stimuli evoked at least twice as many spikes per presentation as did elongated stimuli. A final category of cells failed to meet the criteria for orientation selectivity but responded preferentially to movement in one direction. Direction selectivity tuning was usually quite broad (halfamplitude width 90 ~ or more). These units were called "pure direction selective" (Blakemore et aL, 1972). Transitional unit types with characteristics of more than one category were observed and other units seemed not to fall into any of the designated groups. These units were termed "unclassified" and are reported separately.
Results I. N o r m a l R e a r i n g R e s p o n s e c h a r a c t e r i s t i c s o f v i s u a l c o r t e x u n i t s in n o r m a l l y - r e a r e d cats h a v e b e e n d e s c r i b e d in d e t a i l b y o t h e r i n v e s t i g a t o r s ( B i s h o p , C o o m b s a n d H e n r y , 1971; Hubel and Wiesel, 1962; Singer, Cynader and Tretter, 1975). Our data a r e in g e n e r a l a g r e e m e n t w i t h t h e s e r e s u l t s a n d p r o v i d e a n a p p r o p r i a t e c o m parison group for our experimental animals. Table 1 summarizes our results f o r t h e n o r m a l l y - r e a r e d cat. O u r o b s e r v a t i o n s o n c o m p l e x cells in n o r m a l l y - r e a r e d cats a r e h o w e v e r r e l e v a n t to t h o s e o f t h e " r e c o v e r y " g r o u p a n d a r e p r e s e n t e d h e r e in m o r e d e t a i l . W e h a v e s u b d i v i d e d c o m p l e x cells i n t o t w o t y p e s d e p e n d i n g o n t h e s t r e n g t h o f
Function in Cat Visual Cortex
143
Table 1. Illustrates the distribution of orientation and direction selectivity among the various clas-
ses of cortical neurons in normally-reared cats and in light-deprived cats (LD groups). The total number of cells is given at the end of the caption and the values in the table are percentages. The percentage of units in each subgroup for the light-deprived cats is given in parenthesis. Normal cat (N = 594), LD cat (N = 103) Oriented units
Directional units
Oriented Oriented non-directional directional Simple 4 (0) Complex 4 (2) Oriented unclassified 3 (0) Hypercomplex 2 (0) Pure direction selective Non-oriented stopped Non-oriented concentric Non-oriented general Totaloriented
3l (6) 24 (4) 13 (0) 2 (0)
83 ( 1 2 )
Non-oriented units
Non-oriented directional
Non-oriented non-directional
9 (1) 3 (1) 1 (7) 2 (17) 20 (62) Totalnon-oriented 17 (88)
Total directional 82 (I 1) their summation along the appropriately oriented line. One type of complex cell responded only to moving slits, and small moving spots and irregularlyshaped objects were ineffective stimuli. The majority of the complex cells encountered however, gave vigorous responses to small spots and other irregularly-shaped objects and responses to these stimuli were often as strong as to the best oriented slit which we could find. The response of a direction-selective complex cell of this second type are displayed in Fig. 1. The left hand side of the figure illustrates the responses of the unit to slits of various orientations. As can be seen, there is a vigorous response to the vertical slit, but the horizontal slit evokes no response as it crosses the receptive field. The responses to small spots moving leftward however, are just as vigorous as those to the appropriately oriented slit. Moreover, the range of directions of movement over which the unit responds is much greater when spots are used. Clear responses can be observed to upward and downward movement of the small spot. We have observed vigorous responses to short slits of the "worst" orientation, provided that the slits are "stopped" in length and completely confined to the activating region of the receptive field. These units thus behave like hypercomplex ceils when confronted with stimuli of inappropriate orientations. At the best orientation, increasing the slit length does not result in more vigorous responses, but there is also no inhibition of the response to elongated contours. The responses to the best oriented slit and to the spot are therefore equal. At the other orientations, the units respond only to short stimuli. Transition units between the summating and non-summating type of complex cell can also be observed. In these units one observes an increment in response as the best oriented slit is made progressively longer, but responses to small spots are often quite vigorous.
144
M. Cynader et al.
Slits
,
