As part of a larger experiment, 58 participants worked for a day under one of two lighting designs. The first design used ceiling-recessed parabolic luminaires only; the second design employed the same parabolic ambient lighting, with the addition of an angle-arm task light. Participants had no control over the lighting until the afternoon when they were offered dimming control over the ambient parabolic lighting; participants with task lighting were also permitted to move the arm location. During the day participants performed a variety of simulated office tasks, as well as completing a number of questionnaires on mood, satisfaction, and discomfort. There was no main effect of lighting design on questionnaire outcomes; however, task lighting was associated with performance improvements on some tasks. Interestingly, the provision of a task light did not lower participants’ preferred ambient light output. 

Reference

Click here to see the pdf