212 



VISION 



performance is depicted in Figure 5. Both discrimination tasks were learned 

 within 40 training sessions or fewer, with the stripes discrimination requiring 

 slightly more trials for each shark. Although subject NS-185 did not reach 

 criterion on BW until session 28, it performed significantly above chance by 

 session 15, according to a "runs" analysis of successive correct responses 

 (p < 0.01; Runnels et al. 1968). Histological analysis revealed that the rostral 

 third of the telencephalon, including portions of the lateral olfactory area, 

 was severely damaged in all three subjects. 



— B_W* 



... - H V 



10 20 10 



TRAINING SESSIONS 



Figure 5 Postoperative learning curves for 

 nurse sharks with bilateral anterior 

 telencephalic control lesions. Performance 

 is plotted as a function of the number of 

 trials after starting each task, BW preceding 

 HV in each case. 



The sharks with more posterior telencephalic lesions exhibited varying 

 degrees of visual dysfunction, depending on the extent of damage done to 

 the central telencephalic nuclei. One of the sharks had a lesion that extended 

 only slightly more caudally than the control lesions and did not disrupt the 

 central nucleus (except for its rostral pole) or its pathways. As shown in Fig- 

 ure 6, the performance of this shark (NS-190) did not differ much from that 

 of the controls. Subject NS-189 performed more poorly on the BW task, 

 developing position habits that had to be corrected during training. Also, in 

 spite of some obvious positive transfer, it was unable to reach criterion on 

 the HV task in the limited time available. The lesion in this shark was located 

 farther caudally than that in NS-190 and entirely destroyed the left central 

 nucleus and portions of the right one, transecting both the afferent thalamo- 

 telencephalic tract and the efferent tractus Pallii on the right side. 



The other two subjects in this group were unable to learn either discri- 

 mination task, despite the lack of any motivational impairment. The results 

 of their training are shown in Figure 7. The progressive decrease in median 

 correct response latency indicates that NS-186 successfully learned the 

 general response requirements of the training procedure and could correct 

 errors quickly; however, its performance on the BW discrimination task con- 

 tinued to vary about the chance level throughout the 522 trials. Near the end 



