VISUAL SYSTEM: STATE OF THE ART 81 



Cervetto and MacNichol (1971), using the amino acids L-Na aspartate, 

 L-Na glutamate, and GAB A, recorded the intracellular responses from the 

 horizontal cells of the skate retina. In concentrations of 20 mM the horizon- 

 tal cell membrane potential depolarized until finally the response to light 

 stimuli disappeared. Of the three chemicals tested, only aspartate increased 

 the amplitude of the hyperpolarizing response to light. Glutamate and 

 GABA progressively depolarized the cell membrane and decreased the hyper- 

 polarizing response to light. These results are consistent with the hypothesis 

 of transmitter release in the dark. 



Behavioral Studies of Vision 



Psychophysics— The first behavioral experiment using visual cues for 

 conditioning a shark was that of Clark (1959). She rewarded adult lemon 

 sharks for pressing a submerged plywood target with their snouts. Under 

 these operant methods the animals eventually associated the target with food 

 and ultimately became conditioned to discriminate between a series of tar- 

 gets, indicating their choice by bumping the "correct" stimulus. Since then 

 several studies using the methods of operant conditioning have been accom- 

 plished (see Graeber's article elsewhere in this volume). Among these have 

 been Wright and Jackson (1964) and Aronson et al. (1967), who trained 

 lemon and nurse sharks, respectively, on a visual discrimination task. 

 Aronson et al. first demonstrated that sharks are capable of discriminating 

 targets of differing brightness under an environmentally controlled situation, 

 but neither study quantified the visual results. In addition Aronson et al. 

 showed that sharks learn brightness discrimination about as rapidly as tele- 

 osts and rats under similar conditioning situations. 



Tester and Kato (1966) attempted to condition black tip, Carcharhinus 

 melanopterus, and gray sharks, C. menisorrah, to discriminate between vari- 

 ous targets. Unlike the previous studies which used food reward, their 

 method involved avoidance conditioning. As a result of their somewhat 

 unorthodox procedures, conditioned responses were erratic and led to an 

 imprecise criterion of discrimination. In addition, their subjects were unable 

 to consistently discriminate between more than 50% of the targets, which 

 included such tasks as square vs triangle, gray vs yellow, and bright vs dim. 

 Because of this, Tester and Kato's (1966) conclusions were very tentative. 



Qualitative species differences to the training situation were reported, the 

 black tips being described as more erratic. However, the subjects readily 

 discriminated between rectangles oriented at 90° to each other. Other tests 

 of form discrimination were mostly negative. In view of Graeber's results we 

 must accept Tester and Kato's suggestion that the apparently poor form 

 vision of sharks actually reflected problems related to the experimental 

 methods. 



Results of a brightness discrimination test by Tester and Kato indicated 

 that the subjects could distinguish between gray targets differing by only 

 two Munsell units; color-vision tests were inconclusive. 



