PROCEEDINGS OF THE HUMAN FACTORS SOCIETY-3lst ANNUAL MEETING-1987

The purpose of this study was to demonstrate the utility of the Criterion Task Set (CTS) as a method for personality theory testing. Subjects in a large CTS standardization study were administered the Sensation Seeking scale and the Stimulus Screening scale, two personality dimensions based theoretically on perceptual or biological processes that are believed to mediate task performance. Results indicated that high sensation seekers respond faster, but not necessarily more accurately, than low sensation seekers to central processing tasks. N o differences were found for input/perceptual or motor/output tasks. Also, no differences were found between screeners and nonscreeners for any CTS tasks. The results of this study suggest that the CTS can be used profitably by personality researchers to test the basic assumptions of the theories of some personality dimensions.


INTRODUCTION
The Criterion Task Set (CTS) represents one effort to develop a standardized workload assessment task battery that will aid in the design and operation of complex systems { S hirigledec ker, 1984).
T h e CTS has the distinction of being one of the few, if not the only, task battery based on current theoretical models of human information processing.
The battery is composed of nine tasks, with each task designed to assess one of three primary stages of processing (perceptual/input, central processing, and motor/output).
In addition, there are three workload levels for eight of the nine tasks.
Thus, the CTS provides a theoretically relevant matrix of tasks (stages of processing by workload level) for human performance assessment.
The CTS has been applied a s a test instrument to evaluate the relative sensitivity, reliability, and intrusiveness of a variety of available workload measures.
It has also been used a s a performance assessment battery to evaluate the effects of various stressors on individual components of the information processing system (Shingledecker, Acton, & Crabtree, 1983).

This research was sponsored by the Workload and Ergonomics
Branch o f the Armstrong Aerospace tledical Research Laboratory, United States Air Force, linder Contract F33615-85-0-0514 through the Southeastern Center for Electrical Engineering Education (SCEEE-HER/%-9).
The United States Government i s authorized t o reproduce and distribute reprints for-government purposes notwithstanding any copyright notation hereon. The authors wish t o thank Gary 6. Reid and Dr. Thomas Eggemeier for their interest and support i n the completion of t h i s project.
Anot..ht:r viil~tiblc application o f the CTS i,!; in testing the theories of personality dimensions bc:lieved to be related to task performance ability.
For exrimple, there are several personality variables that are known to relatc to the mmncr in which a person processes imforrrratiort or to the processes related dired.l;y to pcr-forrrraric:o c:npiibility.
Orio such variable, Sexisatiriri Seeking ( S S ) , is : I me>usure of thl: degree to which tt person actively seeks sensory stirnulation.
Since {.he introduction of S S (Zuckcrman, Kotin, Price, & Zoob, 1964) , marly c:cwr.eliitional studies have been performed esf.ij blishing, .for sxamplo, higher illicit drug use, marc jnvolvcment in dnngerous activities, and grtxitcr preference for foods with increased gustritory stimiilation among high, as opposed to low, sensation seekers (Zuckerman, 1979). However, even though the sensation seeking dimension contains considerable relevance for such areas as perception and cognition, few studies of the relationship between S S and basic cognitive or task performance abilities have been reported.
Another relevant personality dimension, Stimulus Screening (SSCR) , is a biologically and perceptually based dimension reflecting the ability to automatically screen irrelevant stimuli during information processing, Stimulus screening represents a heirarchic or patterned approach to information processing (Mehrabian, 1977). Mehrabian (1975) defines those high in SSCR as screeners, or individuals that effectively reduce the complexity or rate of information and therefore evidence a less extreme arousal response.
Nonscreeners , by contrast, are less selective and therefore evidence greater arousal.

PROCEEDINGS OF THE HUMAN FACTORS SOCIETY-31st ANNUAL MEETING-1987
Data collection has recently been completed for a large-scale standardization study that provides a comprehensive data base including CTS performance data, subjective workload assessments, and numerous measures of individual difference variables.
Thus, this data base provides a unique opportunity for combining advances in basic human factors research with theory testing in personality psychology .
The purpose of this paper is to report on the initial examination of the relationship between each of the two personality variables, Sensation Seeking and Stimulus Screening, and CTS performance, Since each of the personality variables examined may mediate human task performance, tach variable presents a uniqut: opportunity to exercise the personality theory testing potential of the CTS. Because this is a n exploratory effort only the most global predictions will be offered.
First, High Sensation Seekers should perform faster and more accmratcly across all of tho (:oi~iporic:ri 1,s of' the CTS battery because of their higher need for stimulation and their higher cwrrc?lutiori with impulsivity.
Screeners, as opposed to nonscreeners, should perform more accurately (and faster) on the perceptual/input component of the CTS battery. The degree to which this differential performance in input/perceptual ability affects other Components of the battery is not easily prce.lic:ted. However, scrccncrs arc: generally viewed as more efficient and should therefore be expected to be more accurate and faster across the other components of the battery, as (:orripareti to noriwrwxwrs.
Briefly, the testing protocol consisted o f re:gularly s<:hr:d71k:d two-horir (X" testing sessions conducted once per day, for nine of ten days, over a two-week period. (The first of the ten days was used for orientation and personnlity Casting.) M u t l i p k : wo.rksl.ril.itxris allowed for the simultaneous testihg of four subjects during each two-hour session. Testing Eight of the nine tasks were performed a t three distinct levels of task difficulty for a total of 25 tasks each day.
A SWAT rating was reported after. the performance of each task. Five days w e r e allocated for training, two days for baseline testing, arid two days for testing under .various stressors.
This paper will summarize the reiationship between two of the personality variables and the performance data from the first day of baseline testing. MANOVA tests evaluating the relationship between S S C R and CTS performance yielded no significant differences (except, of course, for Workload Level).

CONCLUSIONS
The results for the SS groups suggest that high SS individuals a r e generally faster in some cen.tra1 processing abilities, but not in input/perceptual or motor/output ability. This could mean that sensation seekers are simply faster in overall responding in many ways, but no more accurate than low SS types.
These results support the prior finding of higher levels of impulsivity in sensations seekers, but do not suggest a decided performance advantage with regard to accuracy in response.
Of particular interest is the lack of any significant differences between screeners and nonscreeners for the perceptual/input task, Probability Monitoring--a task well suited to test Mehrabian's theory of SSCR. In general, the lack of any significant findings suggests that this dimension does not appear to mediate performance in any robust manner.
It should be noted, however, that the results obtained and conclusions reached a r e limited by the nature of the CTS battery. While high S S types seemed to be faster in central Processing responses, and showed no differences in accuracy, this can only be generalized to the domain of tasks assessed by the CTS.
On the other hand the CTS was designed to assess a fairly wide range of task abilities.
Thus, the results obtained can certainly be viewed as promising and helpful in proceeding toward more sophisticated tests of personality theories.
This study demonstrates the potential value of blending advances in human factors technology and personality psychology to arrive at a method for more adequately exploring the influencc: of personality variables on perf'orrriance. 'it seems particular important that more sophisticated assessrnents of this type be undertaken. Fairly straightforward studies of a single t a s k performance variable in isolation fail to give the wealth in information about the full domain of human lask performance. BY stud yjng rnultiple tasks across workload levels we gain considerable perspoctive on complex himtan-task relationships.