Task Lighting for the Elderly

TL;DR: In this paper, the authors used the Numerical Verification Task (NVT) as a tool for measuring visual performance and subjective preference measures for 17 different lighting conditions and found that the older subjects were more strongly affected by changes in illuminance and in print quality than were the younger subjects.

Abstract: Introduction Task lighting for the elderly poses a special set of problems to the lighting professional and luminaire designer. Reading is an important activity for elderly persons, one that depends on the condition of the visual system. Since the visual system declines from childhood to old age, reading ability and enjoyment can be increasingly difficult as a person ages. Lighting is one of the factors that affects reading ability and enjoyment. Good lighting can help compensate for a decline in a person’s ability to see, while poor lighting can further diminish a person’s ability to see. With the number of persons aged 65 and over expected to at least double in the U.S. in the near future (see Figure 1), lighting concerns for elderly persons will continue to be an important topic within the lighting industry. The purpose of this project was to study a variety of task lighting conditions in an attempt to identify those that are the most favorable for reading by elders. Building upon previous research, this paper describes an experiment in which 17 elderly subjects completed visual performance tasks and subjective preference measures for 17 different lighting conditions. These lighting conditions combined variations in task illuminance, illuminance uniformity, surrounding reflectance, and lamp correlated color temperature (CCT). Age-related physical changes in the eye are well documented (see, for example, Watson, 1996; Cullinan, 1986; Kline and Scialfa, 1997). These changes cause three overall effects that may affect reading ability and enjoyment for the elderly: a reduction in retinal illuminance, a reduction in contrast sensitivity, and an increase in intraocular scatter that reduces retinal image contrast and increases glare sensitivity. However, the potential impact of these effects on elderly persons while reading has not been studied in detail. These effects informed the selection of lighting variables for the experiment reported here. Although visual performance has been studied fairly extensively, relatively few studies involved older subjects. Blackwell and Blackwell (1971) experimented with a large group of visually healthy subjects with ages between 23 and 68 years old, and measured visual performance on a threshold visibility task. They concluded that large differences existed in visual performance capability both among individual observers of the same age and between the averages of different age groups. Using a supra-threshold task, Smith and Rea (1978) asked subjects to perform a proofreading task under different illumination conditions, and compared visual performance between a group of young adults (ages 18 to 22) and a group of older adults (ages 49 to 62). They concluded that there was a large difference in performance between the two subject groups, and that the older subjects were more strongly affected by changes in illuminance and in print quality than were the younger subjects. The summarized results are shown in Figure 2. Rea’s subsequent work (1981, Smith and Rea 1982, Rea, Ouelette and Kennedy 1985) described a series of experiments that provide a model for careful definition and manipulation of task and illumination variables. In addition, these papers described the development and implementation of the Numerical Verification Task (NVT) as a tool for measuring visual performance; this tool was used in the experiment described in this paper. However, these studies all focused on young subjects. Boyce (1973) produced one of the few attempts to quantify the effect of age for a large number of subjects, and for a number of different tasks and conditions. Subjects in three age groups (16-30 years old, 31-45 years old, and 46-60 years old) performed analytical (Landolt-C ring) and simulated (tracker and conveyor tests) tasks and provided subjective preference ratings for five different illuminances. While age did not affect the performance or preference measures for the simulated tasks, a very significant age and illuminance effect was found for the analytical tasks.