טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentEitan Man
SubjectRethinking 'Learned Helplessness Effects': The Cost of
Exploration
DepartmentDepartment of Industrial Engineering and Management
Supervisor Full Professor Erev Ido
Full Thesis textFull thesis text - English Version


Abstract

Overmier and Seligman (1967) found that dogs exposed to inescapable shocks in one situation later failed to learn to escape shocks in a different situation where escape was possible. They termed this phenomenon 'learned helplessness' (LH). Follow up research confirmed the robustness of LH over species and experimental paradigms, and provided an insightful account for human depression, passivity and other 'LH effects'.

The current research compares two interpretations for this pattern in the context of basic decisions from experience: One interpretation asserts that 'LH effects' are the result of experience with no control. The second asserts that the effects are the result of perceiving exploration as costly thereby converging to a local maximum. Two computerized experiments employing a two dimensional search task were devised to measure the predictive power of each interpretation. The use of an explicit economic incentive structure was instrumental in disconfounding uncontrollability and 'punished exploration' characterizing LH classic experimental designs

Experiment I highlights the convergence to a local maximum (the dominant response) in a stable environment. Experiment II generalizes this finding to a frequently changing world in which the incentive structure varies dynamically and demonstrates the sensitivity of participants to the cost of exploration.

Finally, implications are discussed. LH had become the popular line for studying human ills involving passivity such as depression and chronic unemployment as well as offer interventions to prevent or remedy them. An alternative mechanism producing those ills is described in terms of getting stuck in some local maximum as exploration is costly. Different interventions are advised accordingly.