Control and analysis of eye-movement as operant behavior

2006 Fiscal Year Final Research Report Summary

• Project/Area Number: 17330155
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Experimental psychology
• Research Institution: Keio University
• Principal Investigator: SAKAGAMI Takayuki Keio University, Letters, Professor, 文学部, 教授 (90146720)
• Co-Investigator(Kenkyū-buntansha): MASUDA Naoe Keio University, Letters, Professor, 文学部, 教授 (60118510) ; OMORI Takahide Keio University, Letters, Assistant Professor, 文学部, 助手 (60276392)
• Project Period (FY): 2005 – 2006
• Keywords: eye-movement / operant behavior / choice behavior / operant conditioning / mouse device / visual searching / identical options

Research Abstract

The purpose of this work was to evaluate eye-movement operants in human observation and attention. Toward this goal, we used an eye-tracking recorder to record eye movements while the subject made mouse-click choices among visual stimuli presented on a monitor. We used three procedures: (1) identical alternatives in choice; (2) eye-movement conditioning of traveling speed in a visual search task; and (3) directional eye-movement conditioning in a visual search task. We examined the relationships between choice behavior and eye-movements in Situation (1) where participants chose one of identical twelve alternatives in various arrayed positions using a mouse. The results showed that both choice behavior and eye-movements were not turned toward the lower part of the arrays in the first choice trial. From a fine analysis, it appeared that a function of seeing seemed to change from detecting to directing as trials passed. In Situation (2), we prepared eight alternatives marked from one to eight on a circle with equal distances. Participants were asked choose alternatives in increasing order, and their choices were reinforced by points which were correlated or not correlated with their eye-movement traveling speed. Results showed that the conditioning was successful for six of eight participants. In Situation (3), we used a stimulus configuration similar to that in Situation (2), but Landolt-c instead of figures were put on the alternatives, and conducted operant conditioning of eye-movement direction (i.e., clockwise / counter clockwise). Unfortunately we could not get consistent results because of the problems of controlling programs.