|M.Sc Student||Barnea Dganit|
|Subject||The Influence of the Distinction between Task Components|
on Attention Control Processes
|Department||Department of Industrial Engineering and Management||Supervisor||Professor Emeritus Daniel Gopher|
The present research investigates attention control processes by examining the performance costs of switching between tasks. Switching costs are measured by reaction time and errors. Previous research claimed that, switching costs are caused by the requirements to adapt to the new stimulus, response or computational requirements of the new task. The present research asks whether such a change is mandatory, or similar costs can be obtained when only subjective value of task segments is being manipulated. In the fourth experiments that were run, the subjects performed two tasks alternately. In two experimental groups subjective value of segments was manipulated such that in one groups subjects received extra rewards based upon their total average of an 80 trials block, in which moving the displayed information above or below a centrally located line, after 4-12 trials, created 10 segments. In the second, experimental group, subject was rewarded base upon the 3 best task segments. Thus, only for this group task segments had subjective significance. The results show that the subjective manipulation, which separates between units, had significant switching cost. It was also found that is it possible to organize task units even if consecutive trials very in their features and subjects are required to alternate continuously between tasks. The establishment of task units by subjective value was possible as long as the composing elements shared some common features, most importantly responses. However when alternating task elements are distinguished by their response, it is no longer possible to bind together in a single task unit. A common response component seems to be critical for the ability to use subjective significance as a task integration principle. The main conclusion is that task units are not only defined by their physical or computational properties, but also subjective attribution of values. The influence of the latter is akin to those of physical properties.