JOURNAL OF ANIMAL BEHAVIOR 



Vol. 7 MAY-JUNE No. 3 



THE DISTRIBUTION AND ELIMINATION OF 

 ERRORS IN THE MAZE 



HARVEY CARR 



University of Chicago 



In the solution of the maze problem, animals do not distribute 

 their errors equally between the various cul de sacs; apparently 

 these blind alleys do not offer equal incentives to entrance. 

 Neither do the cul de sacs present the same difficulty in mas- 

 tery; the tendency to enter certain alleys is eliminated much 

 sooner than in the case of others. The principles determining 

 the relative frequency of entrance into the various alleys, and 

 the factors governing the order of their elimination are inade- 

 quately known. 



Vincent 1 has published data concerning the first problem. 

 Hubbert 2 and Vincent 1 have investigated the second question, 

 and Watson, 3 on the basis of Hubbert's results, has contributed 

 to the discussion. 



Watson contends that if food satisfaction is a causal agency 

 in selecting the true path and thus eliminating the cul de sacs 

 the latter part of the maze should be mastered first; hence the 

 cul de sacs should be eliminated in the order of their nearness 

 to the food box. He therefore concludes that food satisfaction 

 can not be regarded as a selective agency because only seven 

 of Miss Hubbert's eighty-four rats eliminated the six errors in 

 the exact order of their spatial contiguity to the food box. 



Vincent. The White Rat and the Maze Problem. Jour. Animal Behav., 5, 

 367-374. 



2 Hubbert. Elimination of Errors in the Maze. Jour. Animal Behav., 5. 



3 Watson. Behavior, p. 268. 



145 



