E. C. Poulton

MEASURING THE SPARE ‘MENTAL CAPACITY ’OF CAR DRIVERS BY A SUBSIDIARY TASK

It is impossible to determine the degree to which a driver is absorbed in his driving by measuring his overt responses directly. However, relatively small changes in his spare ‘ mental capacity ’ can be detected by scoring his performance on a subsidiary task, which has no adverse effect upon driving. This technique is sufficiently sensitive to reveal the higher level of concentration required in a shopping area as compared with that required in a residential area

Monitoring up to 16 Synthetic Television Pictures Showing a Great Deal of Movement

A total of 204 observes watched usually 16 synthetic television pictures for 1 hr. The 8 pictures in the middle 2 columns of the display showed activity inside prison. The 8 pictures in the 2 side columns of the display showed activity just outside prison walls. There were 14 suspicious incidents inside prison, 7 of them serious, and 170 other movements. Just outside prison there were fl auspicious incidents, 1 of them serious, and 5,300 other movements. The observer had to press an alarm button whenever he or she saw a suspicious incident. Suspicious incidents were missed reliably more often when they were in the distance, or of brief duration. When matched for distance and duration, incidents were missed reliably less often inside prison, where there was less other activity, than just outside prison. Inside incidents were missed reliably less often when only 8 inside pictures were watched, than when 4 inside and 4 outside pictures were watched- A telephone call about 10 see before an incident reliably reduced the proportion of detections. If more than 1 picture has to be watched, the maximum desirable number is about 9 when there is a lot of movement. The maximum desirable duration for a spell of work is about 1 hr. Reducing the size of pictures from 40 × 30 cm to 20 × 1.1 cm reliably reduced the proportion of detections. Halving the distance of the observer from the small pictures from 9 to 4-5 ft still left a reliable reduction in the proportion of detections of distant incidents. Observers under 30 yr detected reliably more incidents, and made reliably fewer false detections, than observers over 45yr. When matched for age, prison officers who had previously monitored television pictures were reliably better than prison officers who had not. The experienced prison officers detected reliably more suspicious incidents inside prison than did the civilian observers.

ON INCREASING THE SENSITIVITY OF MEASURES OF PERFORMANCE

This paper illustrates some of the methods which have been used to increase the sensitivity of measures of performance: adjusting the difficulty of the task; saturating the mans channel capacity by giving him an additional task to perform; using un unfamiliar task; measuring variability instead of mean performance: selecting specfic events on which to make measurements; examining component rather than overall measures: and channelling two dimensions of variability into one. Finally it raises a methodological difficulty in comparing the results of performance tests which may differ in sensitivity.

Watching for People and Actions

Abstract A total of 24 police officers and 156 civilians were shown a film of a street scene for 4, 2 or 1 hr. They had to watch for people and actions which had been inserted into the film deliberately. Photographs of the wanted people were mounted beneath the screen. In one condition a short colour film of each wanted person was shown before the main film. Three people were detected reliably more often when they were the only throe people to be watched for, than when they were among 12 people to be watched for. No advantage was gained by reducing the number from three to one. The three people were detected reliably less often when the film and photographs were in black and white than when the film was in colour, their photographs were in colour, and the film was proceded by a short colour film of each person. Stealing was detected reliably more often than the wanted person who did the stealing. Both people and actions were detected reliably more often when they were near the camera than when they were f...

The Interaction of the Loss of a Night's Sleep with Mild Heat :Task Variables

Twelve men performed 3 tasks after I night without sleep, at 38°/32°C (100/90°F), and with the 2 stresses combined, as well as in a control condition. The 3 tasks were tracking with peripheral lights, the 5 choice task, and an auditory vigilance task, in that order. There was a reliable interaction between the 2 stresses, but only at the start of the auditory vigilance task. Here the detrimental effect of the 2 stresses combined was reliably less than the sum of the 2 separate detrimental effects. There were also differences between the 3 tasks in the direction of the change in the interaction over time, and in the time taken to show reliable effects of the stresses.

Asymmetric transfer in within-subjects experiments on stress interactions

A comparison is made between the results of three similar within-subjects experiments on mild heat paired respectively with the arouser low-frequency noise, and with the depressants 1 mg of l-hyoscine hydrobromide, and the loss of one full night of sleep. The comparison shows that the level of performance in the control condition of a within-subjects experiment may depend upon the nature of the second stress. Bias in the results of the control condition may produce a reliable main effect, or change the sign of a main effect from positive to negative. The direction of transfer between the conditions of a within-subjects experiment may be positive or negative, depending upon the order of presentation of the conditions. The sign of the interaction between two stresses may reverse between the first and second halves of the experiment. The reliable main effects may be quite different from those of a comparable separate-groups design. Thus the transfer bias should always be measured in a within-subjects design....

Monitoring 16 Television Screens Showing Little Movement

Altogether 65 observers watched 16 or 24 television screens for L hr. The screens showed mainly still scenes from a prison, with occasional movements lasting 2 to 40 see. The observer had to press an alarm whenever he saw a suspicious incident. During the 1 hr there were 12 suspicious incidents and 72 other movements. Reliably more incidents wore missed when they were brief and indistinct, and when they were scattered over 24 screens instead of over only 16 screens. Having to report all movements, and having to answer questions about movements every 30 sec, also reliably increased the number of missed incidents. Having 24 screens to monitor instead of 16, and having to answer questions, reliably increased the time taken to press the alarm. Incidents were spotted more quickly if there had recently been a movement on an adjacent screen. Sitting nearer also reduced the time taken to press the alarm. The results conform to the principles of human performance derived from simpler experiments on visual vigilanc...

EFFECT OF COLD AND RAIN UPON THE VIGILANCE OF LOOKOUTS

Sixteen men performed lookout duties twice at sea in winter on an open bridge, once in the Arctic (mean temperature 28°F) and once in a mores temperate clime (mean 37F°) in counterbalanced order. The 2 signal sources were separated by an angle of 75° and presented 7 signals each in an irregular order and at irregular intervals during a 30-minute watch. The lookout had to respond as soon as he saw a signal. There were reliably more response times of 20 seconds or longer in rain than in the cold (p < -01). There was a reliable increase in the number of long response times during the watches in the cold (p < -01) accompanied by a mean fall in oral temperature of 12°F.

SEQUENTIAL SHORT-TERM MEMORY : SOME TRACKING EXPERIMENTS

Twelve naval ratings had to reproduce an irregular wavy track after a time lag. Rate of input and duration of delay were varied systematically in different trials, both with and without preview, Accuracy in reproducing the positions of the reversals in direction of the track was found to depend only upon the time lag: the longer the delay the worse the reproduction. Amount of input held in memory did not have a significant effect. These quite different findings from previous experiments were attributed to the reduced chances for rehearsal and to the reduced importance of order-information. A theoretical analysis is included of the independent variables affecting sequential short-term memory,

Skimming (Scanning) News Items Printed in 8-Point and 9-Point Letters

Fifty housewives searched for particular words in paragraphs of text printed in 8-point and 9-point Times New Roman without leading in lines of 2·2 inches. To prove that they had found a word they had to write down the following word. The 9-point. print was scanned 7 per cent faster than the 8-point (p<0·05). The method appears to be more sensitive to small changes in the visibility of print than methods involving comprehension.