Correctness of Kruskal's algorithms for monotone regression with ties

Kruskal has proposed two modifications of monotone regression that can be applied if there are ties in nonmetric scaling data. In this note we prove Kruskal's conjecture that his algorithms give the optimal least squares solution of these modified monotone regression problems. We also propose another (third) approach for dealing with ties.Comments by Dr. J. B. Kruskal have been most helpful.     

Substance P facilitation of memory: effects in an appetitively motivated learning task

Food deprived, heterogeneous strain (HS/IBG) mice were trained on two different discrimination tasks for food reinforcement. In one experiment animals were trained to make spatial discriminations in a T maze. Immediately after training they were given subcutaneous injections of either substance P (1 ng/g) or vehicle. Twenty-four hours later the animals were given reversal training in the same maze. The results showed that substance P-treated animals took significantly longer to acquire the reversal habit than did control mice. In a second experiment, animals were trained to make visual discriminations in a T maze. Immediately after reaching acquisition criterion animals were injected with either substance P (1 ng/g) or vehicle. Different groups of mice were retrained on the same task either 1, 2, 3, or 7 days after original learning. Savings scores were calculated and, at every interval, substance P-treated mice retained the task better than control animals. One interpretation of these data is that substance P-treated mice remembered the original task significantly better than vehicle-injected control animals.     

大脑两半球功能不对称性与儿童学业成绩

本研究调查了 3 1 8名小学一至三年级学生大脑两半球优势情况 ,然后对不同优势半球学生的语文和数学成绩进行了比较。结果发现优势半球为右半球的学生人数较少 ,只有 3 .8% ,明显少于优势半球为左半球的学生 (χ2 =2 3 9.83 0 ,p<0 .0 0 1 )。对不同优势半球学生的语文和数学成绩的比较 ,未发现两者有显著差异 (语文 :F=1 .1 3 0 ,p>0 .0 5 ;数学 :F=0 .5 75 ,p>0 .0 5 )。