The literature on the effects of noise on monitoring performance shows a disappointing lack of consistency in results. The hypothesis of the present study was that task classification in terms of demands made on the observer should reconcile conflicting findings so that generalizations could be made. Therefore, a study was made of the effects of intermittent or variable noise on vigilance experiments with similar task demands. Twenty-one sensory vigilance studies, selected from 98 visual performance experiments, were analyzed in detail. It appeared that, even when studies possess similar task characteristics, they are hard to compare due to the many types and varieties of the noise variables involved and the measures of performance used. Contradictory results remain. It was concluded that we know nothing about the effects of variable noise on sustained attention, despite the importance of this kind of noise for everyday life. Using this detailed analysis as an illustration, it was suggested that disparate task definitions contribute to variable or inconsistent results in vigilance, and that it does not make sense to search for the effect of independent variables on “vigilance.” The usefulness of future reviews of “noise and vigilance,” as well as of “the effects of noise,” and the results of “vigilance tasks” were questioned.

In an attempt to specify the limiting conditions of the taxonomy of vigilance tasks, four tasks differing in memory load and in stimuli employed (sensory or cognitive) were compared. Electrodermal activity and subjective measures were used to determine the investment of effort. The data show that vigilance level and vigilance decrement dissociate. The level seems to relate to effort demand and investment; the decrement seems to be task driven, determined mainly by the type of stimuli used. Tasks employing unfamiliar stimuli showed a decline in sensitivity; "cognitive" tasks employing alphanumeric stimuli did not. Principal components analyses suggest that measures of speed and accuracy may reflect relatively independent systems. Subjective data showed that good performers expand more effort in difficult and complex tasks. Effortful processing seems to prevent rather than induce a decline in efficiency.

No abstract

This report describes a test of the prediction, made by Teichner in 1974, that on visual vigilance tasks dynamic stimuli result in greater performance decrements than do static stimuli. For correct detections and sensitivity there was only a nonsignificant trend in the predicted direction, but for response latency (RT) the prediction was supported. Positional uncertainty within displays did not affect vigilance performance. Teichner's assumption that ocular demand is responsible for the impairing effect of dynamic stimuli is questioned. Depletion of central capacity rather than an ocular effect may explain deteriorating performance. Just as in studies of visual fatigue, the contributions of central and modality-specific impairing effects are difficult to separate.

The effects of intermittent noise and its temporal pattern on visual vigilance performance were investigated using a between-subjects design with a multivariate approach. The conventional way of analyzing the data showed a decrement in sensitivity (A′) with time on task, a rise in false alarms, but no significant effect of noise. However, a detailed analysis demonstrated a facilitatory effect of the noise stimuli, which may be explained in terms of orienting reaction theory. It is suggested that the usual way of analyzing data from vigilance experiments may veil the effects of independent variables, such as noise. Some sex differences in performance were also noted. An additional estimation task provided no support for expectancy theory.