Technical Report 764

DTIC FILE COPT

co Effects of NBC Protective Equipment

and Degraded Operational Mode onTank Gunnery Performance

Millicent H. Abel

ARI Field Unit at Fort Knox, Kentucky

Training Research Laboratory

DTIC-dIELECTE1iFEB 1 01988

U. S. Army

Research Institute for the Behavioral and Social Sciences

October 1987

Approved for puhlic release; distribution unlimited.

8'" 031.ký N=="-

U. S. ARMY RESEARCH INSTITUTE

FOR THE BEHAVIORAL AND SOCIAL SCIENCES

A Field Operating Agency under the Jurisdiction of the

Deputy Chief of Staff for Personnel

WM. DARRYL HENDERSONEDGAR M. JOHNSON COL, IN

TechnicJl Director Comrnand ng

Technical review by

Scott E. Graham Accession ForPaul J. Tremont NTIS GRA&J

DTIC TAB"- Unannounced 0l

Justification-.

By,

Distr'ibution/Availability Codes

Avail and/orDist Special

NOTICES 1"TR U Priryý ributio is rep a1'iikas beefQde by fl%.Please ldress corre-

spa N 'ecoc istributlo epart : S. Awjesea- I titute qr e fbvi laland Social Sciences, ATTN' PERI-POT, 5001 Ei nhower Ave., Alexandria, Virginia 22333 00.

FINAL DISPOSITION: This report may be destroyed when it is no longer needed. Please do notreturn it to the U.S. Army Research Institute for the Behavioral and Social Sciences.

NOTE: The findings in this report are not to be construed as an official Department of the Armyposition, unless so designated by other authorized documents.

UNCLASSIFIEDr-CURI tY CLASS•IICAkTION OF THIS PAGE

REPORT DOCUMENTATION PAGEIa. REPORT SECURITY CLASSIFICATION lb. RESTRICTIVE MARKINGSUnclassif ied

2a. SECURITY CLASSIFICATION AUT4lORITY 3. DISTRIBUTION/AVAILABILITY OF REPORT

2b. DECLASSIFICATION I DOWNGRADING SCHEDULE

"4, PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)

ARI Technical Report 764I,, NAME OF PFIFORMING ORGANIZATION I 6b. OFFICE SYMBOL .?,.NAME OF MONITORING ORGANIZATIONU.S. Army esearch Institute (If applicable) S. Army Research Institute for the Be-Field Unit:--Fort Knox PERI-IK havioral and Social Sciences

6c. ADDRESS (City, State, and ZIP Code) 7b. ADDRESS kCitY, State. and ZIP Code)

Fort Knox, KY 40121-5620 5001 Eisenhower AvenueAlexandria, VA 22333-5600

ea. NAME OF FUNDING /SPONSORING ab. OFFICE SYMBOL 9. PROCIUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION (if applicable)

Be- ADDRESS (City, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS

PROGRAM PROJECT TASK WORK UNITELEMENT NO. NO. 2Q26374 NO. ACCESSION NO.

6.37.44.A A795 3.5.1 .5.1.1.311. TITLE (Include Security Classification)Effects of NBC Protective Equipment and Degraded Operational Mode on Tank GunneryPerformance

12. PERSONAL AUTHOR(S)Millicent H. Abel

13a. TYPE OF REPORT 13b. TIME COVERED 14. DATE OF REPORT (YearMonthDay) IS. PAGE COUNTFinal Report FROM1O/ 8 6 TO 07/87 1'1987 October 69

16. SUPPLEMENTARY NOTATION

17. COSATI CODES 18. SUBJECT TERMS (Continue on revere if necessary and identify by block number)YIELD GROUP SUB-GROUP

I• •TRACT (Cor~tinue on reverse if neqessar nd *d nti ', by block number)

1 gunner s performance in Me WII: Conduct of Fire Trainer (UCOFT) was tested while

wearing Mission Oriented Protective Posture (MOPP) gear under normal and emergency opera-tional mode conditions. A pretest, in both experiments, determined initial performance forpossible use as a covariate on experimental test performance. In Experiment I, the subjectsperformed six different test exercises. Results of Experiment I indicated that MOPP gear andemergency operational mode degraded fire time, percent hits, and aiming error. MOPP gear andnormal operational mode degraded aiming error only. The degradationr in performance was de-pendent on the type of exercise. In Eiperiment II, the subjects performed four different testexercises. Results of Experiment II revealed degradation on all performance measures duringemergency operational mode. No performance decrements were found for MOPP gear and no inter-action between MOPP gear and operational mode occurred. The degradation in fire time andaiming error zaused by emergency mode was dependent on the type of exercise. Multiple correlations between fire time and experience variables were significant, particularly overall timeas a gunner. <4..-..

20. DISTRIBUTION /AVAILABILITY ,.F ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATIONS [gUNCLASSIFIED/UNLIMITED 0 SAME AS RPT. 0l DTIC USERS Unclassified

22a. NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (include Area Code) 22c. OFF'PCE SYMBOLMildicent H. Abel (502) 624-3450 PERI-IK

DD FORM 1473, B4 MAR 83 APR edition may be used until exhausted. SECURITY CLASSIFICATION OF THIS PAGEAll oUher editions are obsolete. UNCLASSIFIED

i

Technical Report 764

Effects of NBC Protective Equipmentand Degraded Operational Mode on

Tank Gunnery Performance

Millicent H. Abel

ARI Field Unit at Fort Knox, Kentucky I

Donald F. Haggard, Chief

Training Research LaboratoryJack H. Hiller, Director

U.S. ARMY RESEARCH INSTITUTE FOR THE BEHAVIORAL AND SOCIAL SCIENCES5001 Eisenhower Avenue. Ale-xandria, Virginia 22333-5600

Office. Deputy Chief of Staff fof PersonnelDepartment of the Army

October 1987

Army Project Number Training and Simulation2Q263744A795

Apptoved for public release. distribution unlimnited.

licy 11awdum

FOREWORD

The Fort Knox Field Unit of the Army Research Inrtitute (ARI) conductsresearch in the areas of Armor training and simulation and soldier perfor-mance. The research is sponsored by the U.S. Army Armor Center and School(USAARMC & S) with the objective of increasing soldier readiness and opera-tional effectiveness.

ARI's Fort Knox Field Unit provides research in identifying perfarmancedeficiencies associated with future battlefield conditicos and extent of im-provement in performance with training. Once deficienutes are identified,training countermeasures can be developed to offset the level of degrada-tion and alleviate operational problems that may be present on the futurebattlefield.

The Command and Staff DeparLment at the Armor School in Fort Knox and theU.S. Army Chemical School in Fort McClelland, Alabama, are proponents of Nu-clear, Biological, and Chemical (NBC) simulation projects. This report on theeffects of NBC equipment and operational mode on tank gunnery performance hasbeen provided to these agencies. The research results were briefed to per-sonnel from the Command and Staff Department, Armor School, and the ChemicalSchool, Fort McClelland, in July 1987. Communication between agencies involvedin NBC-related issues results in the identification of key research questionsFnd answers. Consequently, coordinated efforts more rapidly advance today'sunderstanding of training requirements for the future battlefield.

EDGAR M. JOHNSONTechnical Director

v

EFFECTS OF NBC PROTECTIVE EQUIPMENT AND DEGRADED OPERATIONAL MODE ON

TANK GUNNERY PERFORMANCE

EXECUTIVE SUMMARY

Requirement:

The purpose of this research was to examine performance on the Ml UnitConduct of Fire Tr&iner (UCOFT) by gunners wearing Mission Oriented ProtectivePosture (MOPP) gear, in normal and emergency operational conditions.

Procedure:

The research was conducted in two experiments. Subjects in Experiment Iwere assigned to one of four experimental groups: (a) Group 1--no MOPP, normaloperational mode, (b) Group 2--mask only, normal operational mode, (c) Group 3--mask and gloves, normal operational mode, and (d) Group 4--mask and gloves,emergency operational mode. A pretest assessed initial performance for possibleuse as a covariate on experimental test performance. The pretest was performedwithout MOPP gear and in normal operational mode. The subjects completed sixtext exercises under the conditions corresponding to their assigned group. InExperiment II, a group of subjects that had performed exercises under emergencyoperational mode but without MOPP gear was combined with the previous Group 1,Group 3, and Group 4 from Experiment I. Experiment II was designed to assessthe interaction betwee;i MOPP gear and operational mode. The pretest again as-sessed initial performance and was performed without MOPP gear, in normal opera-tional mode. The subjects completed four test exercises in Experiment II underthe experimental conditions for the assigned group.

Findings:

Results of Experiment I indicate that the combination of MOPP gear andemergency operational mode degraded fire time, percent hits, and aiming error,particularly in long-range target engagements. MOPP gear under normal opera-tional mode conditions degraded aiming error only. The results of ExperimentII revealed degradation on all performance measures by emergency operationalmode. No performance decrements were found for MOPP gear, which did notfurther degrade performance under emergency operational mode. The degradationin fire time and aiming error caused by emergency mode was notably evident inlong-range target engagements. Multiple correlations between fire time andexperience variables were significant, particularly overall time as a gunner.

Utilization of Findings:

The results of this research revealed no overall substantial performancedecrements for subjects wearing MOPP gear. These positive results may be theconsequence of the stress on Nuclear, Biological, and Chemical (NBC) training

vii

in today's Army. Performance degradation did occur when subjects performedunder emergency operational conditions. Extensive training under degradedoperational conditions appears mandatory to offset the performance deficien-cies revealed in this research* For mission accomplishment, soldiers needto be trained during peacetime under all potential conditions faced on themodern battlefield. The UCOFT training matrix includes exercises specificallydesigned for training under various degraded operational conditions. There-fore, UCOFT training should be a critical component in training countermeasuresLo alleviate the level of degradation associated with future battlafieldconditions.

I..

k

viii

EFFECTS OF NBC PROTECTIVE EQUIPMENT AND DEGRADED OPERATIONAL MODE ONTANK GUNNERY PERFORMANCE

CONTENTS

Page

INTRODUCTION . . . . . .. . . . a 1

EXPERIMENT 1 4

Method . . aaa.... a 4

Results % .. a a.a.••a . . . . .a 8Discussion t aaaaaaaa 1.3

EXPERIMENT II . .. . . . . . . .aa....a. a 19

Method . . . . . . . . . .a a.........aa.... 19Results aa a*a.a.a a a a a a a a a a a a a 20Discussion .. a . .aaa a a a.a . a a a a a a a a a 31

GENERAL DISCUSSION .. . . . . . . . . .a . a .a.a a 31

REFERENCES . .a a a a a a a a. a a a . a a . 35

APPENDIX A. LISTING OF EXERCISES - EXPERIMENT I .a. . . . . . . . . . . . A-iB. INSTRUCTIONS TO SUBJECTS ... .. . . . .. . B-I

C. BIOGRAPHICAL DATA- EXPERIMENT I .. ....... . ... C-1 N

D. BIOGRAPHICAL QUESTIONNAIRE ......... a. . ...... D--1i

E. PRETEST PERFORMANCE - EXPERIMENT I .. .. . . . . . . . . . E-1

F. INTERCORRELATIONS AMONG PERFORMANCE MEASURES -EXPERIMENT I . . . . . . . . . . . . . . . . . . a ... .a. F-I

G. SUMMARY DATA ON PERFORMANCE MEASURES - EXPERIMENT I a.... G-1

H. INTERCORRELATIONS AMONG BIOGRAPHICAL VARIABLES -EXPERIMENT I . . . . . . . . . . . . . . . . . . . . . . . . H-1

I. LISTING OF EXERCISES -EXPERIMENT II .a. .a .. ... .a. I-I

J. PRETEST PERFORMANCE EXPERIMENT II .. E......a..... J-1

K. INTERCORRELATIONS AMONG PERFORMANCE MEASURES -EXPERIMENT II .a. . . . . . . . . . . . . . . . . . . . . . . K-1

ix

CONTENTS (Contiuued)

Page

APPENDIX L. SUIMARY DATA ON PERFORMANCE MEASURES - EXPERU41NT II . . . . L-I

M. IIOCR0PHICAL DATA - EXPRIKENTIt . . . . . . . . . . . . . -

N. INTERCORRELATIONS A1MONG BIOGRAPKICAL. VARIABLES-EXPERIKENT I1 . . . . . . . .° . ° . N-i

":T OF TABLES

Table 1. Experimental conditions for groups in Experiment I . . . . . . . 6

2. Summary data on experimental test performance bygroup for Experiment I i t. . % . . .0 0 . a %..... .. . 9

3. MANOVA summary table for file time an Experiment I ..... . ° 10

4. MANOVA summary table for percent hits on Experiment I % . . . 11

5. MANOVA summary table for aiming error on Experiment I ..... 12

6. Pearson correlations between performance measures onthe pretest and in the experimental conditions andbiographical variables for Experiment I . .. . . . .a 14

7. Results of multiple regression analyses of performancemeasures by the biographical variables for Experiment I . . . 15

8. Experimental conditions for groups in Experiment II t .... * 19

9. Summary data on experimental test performance by groupfor Experiment II .. . . . . . .0 0 .0 % 0 .0 .0 22

10. MANOVA summary table for fire time on b.periment II . ..... 23

11. MANOVA summary table for percent hits on Experiment II ..... 24

12. MANOVA summary table for aiming error on Experiment II ..... 25

13. Pearson correlations between performance measures onthe pretest and in the experimental conditions and

biographical variables for Experiment II ............ 27

14. Results of multiple regression analyses of performancemeasures by the biographical variables for Experiment II . . . . 28

C-i. Biographical summary data by group for Experiment I . . . . . . C--i

x

CONTET (Continued)

Page

Table R.I. Means and standard deviations on pretest by assignedexperimental group for Experiment I ........... L-

1.1. Pearson intercorrelations on performance measureson the pretest and in the experimental conditionsfor Experiment I * 0 0 0 0 * . 0 * " 0 0 0 0 a 0 . . 6 0 S 0 lýi

G.I. Means and standard deviations for fire, time on each

exercise by group for Experiment I ** ** ** %. % % a 0-1

G..Means and standard deviations for fircen htim on eachexercise by group for Experiment I * a a a % * % a * % * a a %Li-

0.3. Means and standard deviations for percent hitsr on eachexercise by group for Experiment II . . % . . . . . , . 0 % . G-5

J.1. Means and standard deviations fonrets biin assign eacexpercisentby group for Experiment I1 . . . . . . 6 0 0 0 . * L-3

NdI. Pearson intercorrlations on birogrmaphcal measribes o

eecsbygopfor Experiment 11 . . . . . .. -2

xi

7Dý Om ---- MO &

inCTS OF maC PROTnCTIVE sQuIPmEN m A) D GaDEDOPERATIOVAL NODS ON TANK GUNNER PUPSMANCE

INTIONCTION

The future battlefield may Involve Nuclear, Biological, and Chemical(NBC) warfare. Soviet military doctrine encompasses the widespread and lib-eral use of chemical warfare, particularly In surprise attacks (Wagner &Gold, 1982). Given the likelihood of NBC warfae, soldiers must be preparedto use protective equipment In order to survive and sustain extended opera-tions In an NBC environment. The concept of MIS0io-Oriented ProtectivePosture (HOPP) originated from the real need for mission accomplishment withas little reduction In combat effectiveness as possible under NBC conditions.NOPP can be raised or lowered through five levels from MOPPO to NOPP4. TheMOPP levels are defined by the mount of protective equipment being worn bythe soldier. The operational levels of NOIP are described In the following,

MOPPO: no equipment worn, protective mask is carriedMOPPI: overgament wornMOPP2: overgarment, overboots wornIIOPP3 - overgarment, overboots, protective mask with hood wornMOPP4: total encapsulation with overgarment, overboots, protective mask

with hood, and gloves

The mask only option exists if the soldier is protected from direct exposureto contamination. However, this optiou depends upon the type of contamina-tion present even if protection it some form of shelter exists.

The protective equipment, while saving lives and reducing casualties,also presents a number of problems and can degrade performance. The problemaencountered include a loss of manual dexterity, limited visual and hearingcapability, fatigue and heat stress, restricted movement and psychologicaleffects (FM 1-102, FM 3-100). The decrement In operational effectiveness hasbeen estimated as high as 501 of the effectiveness without the equipment(Wagner & Gold, 1982). The performance decrement depends upon task complex-ity, the required MOPP level and the weather. As the level of HOPP increases,protection increases. However, operational effectiveness correspondinglydecreases. Therefore, unit commanders have. to carefully evaluate the situa-tion in order to complete the mission while concurrently ordering the appro-priate MOPP level for reducing casualties. The selection of MOPP level isbased on such factors as "the threat, the mission to be accomplished, thevulnerability of friendly units to enemy attack, the weather, anticipatedwork rate, expected reaction time to a chemical attack, donning speed ofprotective equipment, and degradation impact of protective equipmunt"( F1 17 - 17, p . E - 19 ) .

Numerous research projects have Investigated the impact of NBC protectiveequipment on performance (Muza, 1986). The conclusions of the research pro-jests suggest a greater or lesser degree of performance decrements when sol-diers are wearing NBC equipment. A research report examining the effects ofwearing the 1417 protective mask on combat (performance) tasks found less than

M, pr1

101 average lose due to the mask. The greatest degradation was found invoice coinm'nication with a decrement of 202 or more when compared to unmaskedsubjects (Nontaque, Baldwin, & NeClure, 1959). A more ecnt Investigationon the NI7 protective mask also found decrements in verbal communicationalong wth restricted vision (Barnes, runo, Hulas, Harrab, Rickay, Nerkey,Randall, & Shoemaker, 1983).

Visual limitations as a result of wearitg the protective mask have beenreported in several research projects (Cox & Jeffers, 1981; Mutsa, 1986).Degradation in visiai wbe& wearing the protective mask is a function of sev-eral factors includinga (a) restrictions in the visual field of view, (b)reduced visual acuity, sad (c) altered space and distance perceptiou, (Musa,1986). The visual field of view is restricted because the leos In the masklimits the mount of peripheral vitoln that would exist without the mask.Musa (1986) suggests that the decrements associated with the limitation inthe visual field depends upon the specific task end conditions. ". . . tasksrequiring a large visual field are degraded by mask wear whereas tasks util-iuing a mall visual field might not be affected" (Musa, 1986, p. 30).

Visual acuity has been examined by testing subjects on tracking a targetat a constant rate across a visual field while the direction of the targetvaried randomly (Wiley, leabr, Chiow, & Holly, 1977). Wiley, et al. (1977)found that for those wearing a meek, the target angular site had to be in-creased up to 382 over the no-mask condition to achieve a 951 detection rate.The results of Wiley, et al. indicated that the mask interfered with theability to detect and track a rapidly moving target. These results wereextended by Robrick and Sleeper (1986) when they measured response times tovisual stimuli. The response times for detecting visual signals were aig-nificantly longer for those subjects wearing the mask in HOPP4. The resultsof these two projects suggest the protective mask can seriously impose visuallimitations.

Previous research has indicated that the protective rubber gloves cause asubstantial degradation in manual dexterity (Jauson& Jepson, 1982; Johnson &Sleeper, 1986). Manual dexterity includes: (a) fine motor response, (b)fine motor manipulation, and (c) fine motor strength (Ramirez, Shew, Felt, &Rayle, 1986). Johnson and Sleeper (1986) investigated the effects of wearingthe mask, hood, and gloves in various combinations on one-handed andtwo-handed tests of manual dexterity. Their results indicated a substantialeffect on performance for the protective gloves with or without wearing the

Al mask. These results were also supported by Rauch, Witt, Banderet, Tausou,and Golden (1986). Their results revealed significant performance decrementson paper and pencil tests of math computation for those subjects wearinggloves. The mask had no significant effect by itself.

King and Frelin (1982) investigated the performance of medical special-ists on nine medical tasks in fatigues and HOPP4. Their research focused onthe effects of protective gloves on manual dexterity. Subjects in HOPP4 per-formed all the tasks significantly slower (302 to 552 on the average) thansubjects in fatigues. However, King and Frelin (1982) did find substantialperformance improvement over 6 days of practice when in MOPP4. These resultsindicate the substantial impact of the protective gloves on performance andyet the Importance of regular training on tasks while In MOPP gear to reduceI 2

the impact (Bensel, 1980). Several of the aforementioned research projectshave found a significant effect foi NBC equipment on speed or rate of taskcompletion while no differences were found on accuracy (Fine & Kobrick, 1986;Kobrick & Sleeper, 1986; kauch et a:..: 1986).

While soldier's performance in MOPP gear has been examined, relativelylittle data e)ist concerning the performance of Armor crewmen under NBC con-ditions. Existing research has focused on the physiological aspects of heatstress and fatigue resulting from the operational MOPP4 level with completeensemble rather than performance variables such as the psychomotor tasksrequired in tank gunnery (Carr, Kershner, Corona, & Jackson, 1980). Perform-ance degradation in tank crews has been examined in terms of stay time, thelength of time in proLective posture that a unit remains effective beforeexperiencing casualties because of heat stress (Rakaczky, 1981). -Work/restcycles have been developed for crewmen under different MOPP levels dependingupon temperature and humidity (FM 3-4).

Past research indicates the importance of assessing the effects of NBCprotective equipment on Armor crewmen performance. Tank gunnery requiresvisual acuity, manual dexterity, and verbal communication for successful taskperformance. Because research already suggests performance decrements inthese three areas, it seems imperative to determine the extent of degradationin gunnery performance. Performance decrements are also possible due toequipment failures resulting from battlefield conditions. Therefore, assess-ing deficienciet, associated with MOPP gear and degraded operational. condi-tions is warranted. Once the deficiencies have been determined, the amountof training required to counter the degradation can be ascertained.

Ren;earch Objectives

The Unit Conduct of Fire Trainer (UCOFT) is the most sophisticated simu-lator that is available for training tank commander/gunner pairs. Perform-ance degradation resulting from the use of MOPP gear under variousoperational conditions can be simulated on the UCOFT. The objectives of thisresearch project are:

(1) to quantify gunner performance on the UCOFT when wearing the protec-tive mask and gloves under normal and emergency operational mode and

(2) to suggest training countermeasures to alleviate possible performancedeficiencies.

The research wae designed to assess the effects of the protective maskand gloves on visicn and manual dexterity. The operational MOPP4 levelwhich includes the mask and gloves was not simulated. The overgarment,overboots, and hood :.uired at the MOPP4 level were not used in order tocontrol for possible heat stress and fatigue. Therefore, the results forMOPP gear in this research are only based on the effects of mask and glovesand cannot be generalized to the actual operational MOPP4 level.

3

EXPERIMENT I

Method

Subjects

The subjects were 48 Ml qualified gunners from the Ml New EquipmentTraining Team (NETT) at the U.S. Army Armor Center (USAARMC) at Fort Knox.Four M1 qualified tank commanders (TCs) from the Ml NETT Team assisted intesting gunner performance by serving as tank ccmmanders.

Apparatus -I

The MI Unit Conduct of Fire Trainer (UCOFT) was used as the medium forLweasuring gunner performance. The UCOFT is a high fidelity gunnery trainerwhich employs computer-generated target scenarios and a realistic crew com-partment. The majority of controls and switches used by the gunner and TC inthe Ml tank are simulated in the UCOFT. Target exercises on the UCOFT samplenumerous engagement conditions such as weather and visibility conditions,tactical

situations, and equipment

operational readiness.

The Instructor/

Operator (I/0) controls the trainer from an external station. The I/O's

station is equipped with visual displayn of the tank commander and gunner's

sights. The information dimplayed includes target type, reticle used, opera-

tional status and the running time for target appearance with the directiorand magnitude in degrees of the target from the retici.e. The I/O can observewhat the TC/gunner pair view through their sights and can monitor and controltheir progress through the exercises. A computer print-out details perform-

ance during each engagement within each exercise by recording measures such

as number of target hits, number of rounds fired, and target identification

To simulate Nuclear, Biological, and Chemical (NBC) conditions in the

present experiment, the M25/M25A1 tank protective mask and the chemical-pro-

tective glove set were used as the Mission Oriented Protective Posture (MOPP)gear. The hood, overgarment, and overboots were not used in the project topreclude the possible physiological effects of heat stress and fatigue.

UCOFT Exercises

All exercises employed in the present experiment included single targetengagements with day unlimited visibility. The targets were either T72tanks, helicopters, personnel carriers, or trucks. Only main gun engagementswere used. The gunner could fire only one round of ammunition per targetengagement. The gunner had 20 seconds after full target exposure to kill thetarget before the engagement was terminated by the computer. In stationaryown tank engagements, the engagement began in a defensive posture (turretdown position). The gunner's tank could be exposed (hull down position) foronly 15 seconds before the computer simulated its destruction. (See AppendixA for listing of specific exercises.)

4

Procedure

Prelimitary training. The 110s and TCU were specifically trained an allexercises used in the experiment to standardize procedures in the pretest andexperimental conditions. The TCs were trained to issue the appropriate firecommand corresponding to each engagement using doctrinally correct Ml conductoi fire procedures. This procedure standardized the fire commands for eachsubject. The TCs were also trained to lay the reticle on a predeterminedlandmark (i.e., house, rock, tree, etc.) for each engagement before targetpresentation. This was done to ensure the same starting point for all sub-jects in each engagement. This procedure increased the standardization oftarget acquisition time measures (see Smith & Graham, 1987). The landmarkswere approximate~ly five mils ei~ther left or right of the -pot-ential -targetappearance. The targets were within the gunner's 3X sight picture when theyappeared. Additionally, the I/Os were trained in the following duties:

(1) to identify the exercise to the TC and gunner.

(2) to give instruc~tions explaining the exercise to the subject.

(3) to direct the TC to the landmark before target appearance.

(4) to punch in the gunner's target identification response.

(5) to act as the driver by moving the tank In and out of defiladeduring stationary own tank engagements.

(6) to act as the driver by stopping and starting the tank duringmoving own tank engagements in emergency mode.

(7) to print out all performance measures and call up the next exer-c! se.

Group assignment. Subjects were randomly assigned to one of four experi-mental conditions defined by level of MOPP and operational mode. The fourexperimental groups were: (a) Group 1:* No MOPP, normal operational mode,(b) Group 2: Mask only, normal operational mode, (c) Group 3: Mask andgloves, normal operational mode, and (d) Group 4: Mask and gloves, emergencyoperational mode (see Table 1).

Instructions-Pretest. Subjects were read instructions explaining thenature of the testing procedures and the UCOFT (see Appendix B). The rangeof experience on the UCOFT was considerable (see Appendix C for biographicalsummary data). Therefore, initial performance was measured for possibleinclusion as a covariate in the analyses. A pretest was used which consistedof 20 engagements representing the different types of exercises used in theexperimental conditions. Subjects were not in MOPP gear for the pretest andperformed under normal operational mode. Subjects were allowed to ask anyquestions about the UCOFT or the experimental procedures during this pretestperiod. They were told that no questions would be Answered or assistanceprovided by the I/0 or TC during the experimental conditions. Subjects werethen given a 10 minute break during which time they completed a biographicalquestionnaire (see Appendix D).

Table 1

Experimental Conditions for Groups in Experiment I

MOPP Gear Operational Mode

Normal Emergency

No MOPP Group 1

Mask Group 2

Mask and Gloves Group 3 Group 4

Experimental conditions. After the break, those subjects in the groupsrequiring MOPP gear donned the appropriate equipment and proceeded with thetest exercises. TCs were not in MOPP gear. All subjects completed six testexercises on the UCOFT in two sessions of three exercises each. The twosessions were separated by a ten minute break to control for the effects offatigue. The six test exercises were counterbalanced within each group. Thesix exercises were the following:

(1) stationary own tank, long range stationary targets.

(2) moving own tank, short range stationary targets.

(3) moving own tank, long range stationary targets.

(4) stationary own tank, short range moving targets.

(5) stationary own tank, long range moving targets.

(6) moving own tank, short range moving targets

For Groups 1, 2 and 3, the exercises required precision gunnery using thegunner's primary sight (GPS). For Group 4, the conditions were degraded toan emergency mode. Subjects were informed that the computer, laser rangefinder, stabilization system, and the GPS were nonfunctional prior to com-mencing the exercises. Switches on the UCOFT were used to change operationalreadiness. The switch configuration was an alternative to employing UCOFTexercises specifically designed to simulate the degraded conditions. Byusing the switches on the UCOFT, the exercises then remained identical to theother three groups except for operational mode. The Mode switch was placedin Off; the Range switch was placed in Safe; the Thermal Mode switch was inthe Off position; and the Filter/Clear/Shutter switch was placed in Shutter.With the GPS nonfunctional, the TC had to use the 50 Caliber sight for makingthe initial lay on the landmark. In Group 4, under emergency operational

6I

mode, short range engagements requJrad Battlesight techniques. For lougrange targets, the actual target range was included in the fire command. Thegunner had to use the gunner's auxiliary sight (GAS) to apply manual lead tomoving targets. For moving own tank exercises, the gunner had to engage thetarget from a short halt.

Performance Measures

Performance measures of fire time, percent hits and aiming error wereused in both the pretest and the experimental conditions. Measures extractedSfrom the UCOFT print-outs included: (a) average fire time, the elapsed timein seconds from full target -exposure-to when the gunner fires, (h) Uumber oftarget hits, (c) number of targets presented, (d) average azimuth error inmils, and (e) average elevation error in mils. The rounds of ammunitiongenerated in the UCOFT have a greater or lesser degree of dispersion. Atarget hit that results from a dispersion round may or may not be an actualhit. Therefore, this type of engagement was deleted from all exercises. Amiss that results from a dispersion round that would have been a hit other-wise was counted as a hit.

Percent hits was computed as a measure of accuracy for each exercise fromthe number of hits and number of target presentations. Aiming error wascomputed and equaled the root mean square (RMR) of azimuth and elevationerror (Smith & Graham, 1987), RMS - -.%J azimuth errorz + elevation err4orz.Aiming error is an index of the distance in mils from the reticle to thecenter of target mass at time of round impact.

Biographical Measures

All subjects completed a biographical questionnaire. The biographicalmeasures were: (a) total time in the military, (b) total time in Armor, (c)total time as a gunner, (d) self-reported General Technical (GT) score fromthe Armed Services Vocational Aptitude Battery (ASVAB), and (e) number ofhours of experience on the UCOFT. Time in military, time in Armor, time as agunner, and hours on the UCOFT were used as measures of experience. The GTscore was considered a measure of aptitude.

Experimental Design

The experimental design was a 4 x 6 repeated measures design with fourgroups and six different exercises. Multivariate analysis of variance onrepeated measures was used with post hoc comparisons on all significAnt maineffects and interactions. Pearson correlations were obtained between per-formance measures and the biographical variables. Average performance meas-ures across exercises in the experimental conditions were used in thecorrelational analyses. Multiple correlations between performance measuresand biographical variables were obtained from multiple regression analysesusing a hierarchical forced entry procedure (Cohen & Cohen, 1983). The expe-rience measures of time in military, time in Armor, time as gunner, and hourson UCOFT were entered into the regression equation as a functional set of

7

variables. Because the sample included subjects with a great amount of expe-rience, experience was hypothesized as having a larger effect on performancethan GT sctore. Thus, the set of experience variables was always enteredfirst into the equation and the GT score was entertd last.

Result&

Subjects were randomly assigned to one of four experimental groups:

(1) Group 1: No MOPP, normal operational mode.

(2) Group 2: Mask only, normal operational mode.

(3) Group 3: Mask and gloves, normal operational mode.

(4) Group 4: Mask and gloves, emergency operational mode.

All groups received a pretest consisting of representative exercises from theexperimental conditions. The pretest was identical for all subjects. Sub-jects were not in MOPP gear for the pretest and performed under normal opera-tional mode conditions. After the pretest, subjects performed six differentexercises under the experimental conditions according to assigned group.Performance measures of fire time, percent hits, and aiming error were usedin both the pretest and in the experimental conditions.

Pretest Performance

A one-way multivariate analysis of variance (MANOVA) was performed onpretest performance measures to assess possible group differences in initialperformance (see Appendix E for summary data on pretest performance). Asignificant group effect was found for aiming error (F(3,44)=3.99; pu0.13).Fire time, percent hits, and aiming error were intercorrelated (see AppendixF). Therefore, all three pretest performance measures were subsequently usedas covariates in the analyses of experimental test performance.

Experimental Test Performance

Multivariate analysis of variance (MANOVA) on repeated measures withconstant covariates was used to determine if significant group differencesexisted on experimental test performance (see Table 2 for summary data).(See Appendix G for complete summary data on performance measures by exe.-cise.) MANOVA summary tables present complete results of the analyses (seeTables 3, 4, and 5). As detailed in the MANOVA tables, the overall groupmain effect was significant for each performance measure. Results of the posthoc comparisons between groups indicated a significant difference for Group 4(Mask and gloves, emergency mode) on each measure. The only significant dif-ference for the groups in MOPP gear under normal operational mode (Group 2and Group 3) was on aiming error (see Table 5).

8

Table 2

Summary Dtta on Eperimental Teat Performance by Group for Experiment I

Average Average AverageGroupa Fire Time Percent Hits Aiming Error

1. No MOPPNormal Operational ModeM 13.33 78.94 1.17SD 1.84 9.42 0.24

2. Mask OnlyNormal Operational Modeii 14.46 79.66 1.27

SD 2.06 4.41 0.29

3. Mask and GlovesNormal Operational Mode

M 13.58 80.51 1.23SD 1.96 3.83 0.22

4. Mask and GlovesEmergency Operational ModeM 15.37 65.01 1.56

S1.69 8.56 0.26

Note. Fire time is the elapsed time (seconds) from full target exposure towhen the gunner fires. Percent hits - (number of hits/number of tar-gets presented) x 100. Aiming error (mils) - square root [(azimutherror) 2 + (elevation error) 2].

a. - 12 per group.

The MANOVA tables indicate a significant exercise main effect for eachperformance measure. Post hoc comparisons between exercises revealed a sig-nificant effect for long range targets on fire time and percent hits in allgroups with lower performance in long range target exercises compared toshort range target exercises for both variables (see Tables 3 and 4). Thecomparisons also indicated significantly lower performance on each measure inthe stationary own tank/moving target exercise compared to the moving owntank/stationary target exercise. The interaction between target range andthe stationary own tank/moving target exercise and moving own tank/stationarytarget exercise was significant for all measures. Summary data for theindividual exercises indicate for fire time, range had a more substantialeffect in the moving own tank/stationary target exercise versus thestationary own tank/moving target exercise. On the other hand, for percenthits and aiming error, the reverse was true. Range had a greater effect inthe stationary own tank/moving target exercise versus the moving owntank/stationary target exercise.

9

Table 3

MANOVA Summary Table for Fire Time on Experiment I

Source (Adjusted) . df F p level

MAIN EFFECTS:

Group 3,43 10.20 .000Exercise 5,40 67.46 .000

INTERACTION:

Group x Exercise 15,110.82 2.45 .004

POST HOC COMPARISONS:

GroupGroup 4 versus Group 1, 2, 3 1,43 25.33 .000Group 2, 3 versus Group 1 1,43 0.82 ns

Target RangeExercise 1, 3, 5 versusExercise 2, 4, 6 1,44 193.69 .000

Stationary Tank/Moving Target -.

Moving Tank/Stationary TargetExercise 4, 5 versus Exercise 2, 3 1,44 115.84 .000

Target Range x Stationary Tank/MovingTarget-Moving Tank/Stationary Target 1,44 30.60 .000

Group 4 versus Group 1, 2, 3 x Exercise 5,40 4.62 .002

Group 2, 3 versus Group 1 x Exercise 5,40 1.85 ns

Group 4 versus Group 1, 2, 3 x Target

Range 1,44 18.35 .000

Group 4 versus Group 1, 2, 3 xStationary Tank/Moving Target-Moving Tank/Stationary Target 1,44 0.08 ns

10

Table 4

MANOVA Summary Table for Percent Hut* on atperiment I

Source (Adjusted) d. p level

MAIN EFFECTS:

Group 3,43 15.39 .000Exercise 5,40 51.36 .000

INTERACTION:

Group x Exercise 15,110.82 2,68 .002

POST HOC COMPARISONS:

GroupGroup 4 versus Group 1, 2, 3 1,43 44.96 .000Group 2, 3 versus Group 1 1,43 0.09 ns

Target RangeExercise 1, 3, 5 versusExercise 2, 4, 6 1,44 38.47 .000

Stationary Tank/Moving Target -Moving Tank/Statiouary Target

Exercise 4, 5 versus Exercise 2, 3 1,44 88.50 .000

Target Range x Stationary Tank/MovingTarget-Moving Tank/Stationary Target 1,44 12.33 .001

Group 4 versus Group 1, 2, 3 x Exercise 5,40 8.01 .000

Group 2, 3 versus Group 1 x Exercise 5,40 0.15 ns

Group 4 versus Group 1, 2, 3 x TargetRange 1,44 0.15 as

Group 4 versus Group 1, 2, 3 xStationary Tank/Moving Target-Moving Tank/Stationary Target 1j44 41.69 .000

I11

Table 5

HAtNOVA Summary Table for Aimin~g Error an Experiment I

Source (Adjusted)_ df F lvl

MIAIN EFFECTS z

Group 3039.08 .000

Exercise 5140 71.53 .000

K- INTERACTION.,

Group x Exercise 15,110.82 2.73 .001

POST HOC COMPARISONS:

GroupGroup 4 versus Group 1, 2, 3 1,43 23.44 .000Group 2, 3 versus Group 1 1,43 3.96 .053

Target RangeExercise 1, 3, 5 versusExercise 2, 4, 6 1t44 0.26 ns

Stationary Tank/Moving Target -

MoviDg Tank/Statimrary TargetExercise 4. 5 versus Exercise 2, 3 1,44 69.53 .000

'target Range x Staticnary Tank/MovingTarget-Moving Tank/Stationary Target 1,44 4.21 .046

Group 4 versus Gro.: 1, 2, 3 x Exercise 5,40 9.41 .000

Croup 2, 3 ve:zcs Group 1 x Exercise 5,40 0.27 ns

Group 4 vers~us Group 1, 2, 3 m* TargetRange 1,44 9.37 .004

Group 4 versu.s Group 1, 2t 3 xStationary Tenk/?1oving Ta.J~et-Moving Tank.fSnztioaary Target 1,44 44.29 .000

L-10

12

B- 01 IN A

A significant interaction between group and exercise was found for eachSmeasure as iliursrated in the tANOVA tables. The .ost. hoc comparisons re-

vealed a significant interaction between Group 4 and txercise on each mas-Sure, No significant interactions were found between the HOPP groups (Group 2

and Group 3) and exerciseo Group 4 d a significantly slower fire time Inlong range target exercises with no difference for Group 4 in the short rangetarget exercises (see Table 3). A significant difference existed for Group 4on percent hits in the stationary own tank/moving target exercise with nodifference in the moving own tank/stationary target exercise (see Table 4).For aiming error, significant effects existed for Group 4 on long range tar-get exercises and the stationary own tank/moving target exercise (see Table5).

Pearson Correlations and Multiple Regression

The relationship among performance measures and the biographical varia-bles were examined using Pearson product moment correlations. Results sug-gest a stronger relationship between biographical variables and fire timethan with the other performance measures. Seven of the ten correlationsbetween fire time and the biographical variables were significant. Only twoof the ten correlations between percent hits and the biographical variableswere significant and four of ten correlations between aiming error and bio-graphical variables were significant (see Table 6).

Multiple correlations between performance measures on the pretest and inthe experimental conditions and measures of experience and CT score wereexamined by multiple regression. A hierarchical forced entry procedure wasused in the multiple regression analyses (Cohen & Cohen, 1983). The experi-ence measure of time in military, time in Armor, time as gunner, and hours onUCOFT were entered into the regression equation as a functional set of varia-bles (see Appendix H for intercorrelations between biographical variables).Because the sample included subjects with a great amount of experience, expe-rience was hypothesized as having a larger effect on performance than the GTscore. Thus, the set of experience variables was always entered first intothe equation. The GT store was entered last. If the overall F statistic forthe set was significant, the Beta weights of constituent variables in theregression equation were examined separately.

The multiple correlation between fire time on the pretest and the func-tional set of experience variables was significant, (R-.59, R2 -. 35,F(4,42)-5.71; p-.O01). Two variables were significantly weighted in theregression equation: time as gunner (t[42]-2.11, pm.041) and hours on UCOFT(t[42]-2.60; p-.013). The negative sign of the t values correspond to thenegative Beta weights which are in the expected direction. The GT score wasnot significantly weighted in the regression equation. The multiple correla-tion between average fire time in the experimental conditions and the set ofexperience variables was also significant, (R-.52, R2 =.27, F(4,42)-3.90;p-.O09). The only significantly weighted constituent variable in the equa-tion was time as gunner, (t[42]-2.59; p-.013), with the negative Beta weightagain in the expected direction. The GT score was not significantly weightedin the equation.

13

Table, 6

Pearson Oozrtolat ion, between ?erformm~es Neesures an the Pretest end In theExperimental Couditioas and Biographical Variables for Uperiment I

Time In Times In Time as CT Hours onmilitary Armor Gunner Score UCOFT

pretest .4022 .3860 -.3129 .2280 -.3483Fire Time ( 48) ( 48) ( 48) ( 47) ( 48)

P-.002 p0.003 p-.015 pa.062 pa.008

Pretest -.1828 -.2483 .1421 .1324 .2478Percent Hits ( 48) ( 48) ( 48) ( 47) ( 48)

p-.107 P-.044 P-.168 p-. 188 P-.045

pretest .1668 .1358 -.1645 -.2388 -.1674Aiming Error ( 48) ( 48) ( 48) ( 47) ( 48)

p-.129 P-.179 p-.132 pe.053 P-m.128

Experiment .3460 .2383 -.4092 -.0183 -.2013Fire Time ( 48) ( 48) ( 48) ( 47) ( 48)

P-.008 P-.051 p-.002 p-.451 Pmm.085

Experiment -.0379 .1141 .2291 .1082 .0880Percent Hits ( 48) ( 48) ( 48) ( 47) ( 48)

p-.399 p-.220 P-.059 p-.2235 p-.276

Experiment -.1383 -.2739 -.3011 -2644t .0352Aiming Error ( 48) ( 48) ( 48) ( 47) ( 48)

p-.174 p-.030 p'm.019 p-.036 P-.406

There were no significant multiple correlat ions between percent hits inthe pretest or experimental conditions and the biographical variables. Therewas no significant multiple correlation between pretest aiming error and thebiographical variables. However, the multiple correlation between average

aimig eror n th exerimnta conditions and the set of experience varia-bles wasn significant, (R~-.46, Rt -.21$ F(4,42)-2.79; p-m.038). The only sig-nificantly weighted constituent variable in the regression equation was againtime as gunner, (t[421-2.43; p-.019) with a negativo Beta weight in theexpected direction. (See Table 7 for details of the multiple regressionanalyses).

14

C4 4

emm

101 4.90

9)4 "49

14 14 'ow

14 60

o1 1 a='b41 .9 ' .. 9.

tn., ;k~ 14 F

0 Nt'.

4.415

I-

0"40t~4(4 0 "44Q

"C4 S

di~ LIN"C6 'At4f~" - C"

r- W45 * S

F-4 V-4

.0.103 3 33.4 q

be4

.9VV4

.00* "4

11-04S S

'4"4 " "4016

Sx A -Afil I jliI M 111111 ii'l'ilkIA ~ ~ l1

44 -H

0 Cr-

44 1-4 00000 00000

.a n .- C4 N %0 4c ci

oo c4c C4 c 4

4.4 00 '-4- 0-0

0 N P 04-NN-4M4 47 4- 4 4T

1- co ONc'J

04 04C

4)

-4. CO

a4 aadaa4)U

04

.0

41.u

93 -r4

.0 0 0 0

o1 1044

,%0 0 w 14

"4 be w 4 4 H4 Z=r

14 HC2$4mt

H44 C 0.4 Aj W - E kC' o WE4&

8.4 * 41 .4) 00 4) E4 c

04 (A

6d4.4 0

E4 ad CzC'

U '4 * * .17

%). scuDsion.

The purpose of Excperiment I was to examine the effects of MOP-P gear and

operational mode on gunner performance. Aiming error was significantly

larger for the groups in MOPP gear, but no significant effects for MOPP gear

were found for fire time and percent hits under normal operational mode con-

ditions. Overall, the results suggest that wearing MOPP gear in a fully

operational tank does not necessarily lead to large performance decrements in

gunner performance. However, when equipment failures force emergency opera-

tional mode conditions, In addition to HOPP gear, a significant performance

drop occurs.

The re~sult Ia also- indicate that -the level --of -gunner -performance decrement

when wearing IIOPP gear under emergency mode conditions may depend upon the

type of target engagement. For example, HOPP gear and emergency mode sig-

nificantly degraded fire time and aiming error in long range target engage-

ments, but did not affect performance in short range target engagements.

Also, in the stationary own tank and moving target exercise, percent hits and

aimiing error were significantly affected by HO1PP gear and emergency mode,

while no differences-existed when the tank was moving and the targets were

stationary. However, in emergency mode conditions when the stabilization

system is nonfunctional, the gunner must wait for the tank to halt before

firing. This factor caused the moving own tank/stationary target exercise to

become a stationary own tank/stationary target exercise for Group 4. In sum,

the results suggest that the effects of degradation on gunner performance may

depend upon situational determinants and performance decrements cannot be

assumed for all target engagements.

The multiple regression analyses revealed a significant multiple correla-

tion between gunner fire time and the set of experience variables, particu-

larly overall time as gunner. The negative Beta weight in the regression

equation corresponding to time as gunner was in the expected direction. This

result was also found for aiming error in the experimental conditions. These

results suggest that when time as gunner increases, fire time decreases,

hence, the faster the gunner's speed at firing a round of ammunition, and

also the gunner's aiming error decreases. Therefore, the level of gunner

experience may greatly affect certain performance measures under degraded

operational conditions.

The present experiment suggest significant effects of MOPP gear and

emergency mode conditions on gunner performance when they occur simultaneous-

ly. Tne present experiment, however, did not examine the interaction of MOPP

gear and o~perational mode. Level of performance degradation may be greater

when combining MOPP gear and emergency mode conditions than when either fac-

tor occurs alone. The interaction between MOPP gear and operational mode is

thus examined ir x' periment II.

18

EXPERIMENT II

Method

In Experiment I, the combination of MOPP gear and emergency operationalmode conditions significantly impaired gunnery performance. In order toassess the interaction between MOPP gear and operational mode, another sampleof subjects was formed. Data from a group of subjects which performedexercises under emergency operational mode on the UCOFT under the same switchconfigurations but without MOPP gear were available from another project(Witmer, in progress). The subjects were also M1 qualified gunners from theM1 NETT Team, but were not subjects in Experiment I. A 2 x 2 experimentaldesign was thus formed. The experimental -groups--were: (a) Group 1: No-MOPP,normal operational mode (from Experiment 1), (b) Group 2: No MOPP, emergencyoperational mode (from Witmer's project), (c) Group 3: MOPP (mask andgloves), normal operational mode (from Experiment I), and (d) Group 4: MOPP(mask and gloves), emergency operational mode (from Experiment I) (see Table8).

Table 8

Experimental Conditions for Groups in Experiment II

MOPP Gear Operational Mode

Normal Emergency

No MOPP Group 1 Group 2

Mask and Gloves Group 3 Group 4

The testing procedures for both projects were identical which includedthe duties of the I/Os and TCs, the pretest for beginning performance, andsix test exercises. The single discrepancy between the two projects occurredin the test exercises. Only four of the six exercises were the same in bothprojects. Therefore, only those four exercises which were the same wereincluded in the data analyses. The other two exercises were omitted from theexperiment. The four exercises were:

(i) Lioving own tank, short range moving targets.

(2) stationary own tank, loag range stationary targets.

(3) stationary own tank, short range moving targets.

(4) stationary own tank, long range moving targets.

19

All exercises were single tank engagements with day unlimited visibility.Only main gun engagements were used and the gunner could fire only one roundof ammunition per target engagement (see Appendix I for listing of specificexercises). See Method section in Lcperiment I for further details of test-ing procedures and explanation of exercises and conditions. The performancemeasures and biographical measures remained the same as in Experiment I.

Experimental Design

The experimental design was a 2 x 2 x 4 repeated measures design with twolevels of MOPP gear, two levels of operational mode, and four exercises.Multivariate -analysis of variance on repeated measures- was used-- with- ost hoc....comparisons on all significant main effects and interactions. Pearson corre-lations were obtained between performance measures and the biographical vari-ables. Average performance measures across exercises in the experimentalconditions were used in the correlational analyses. Multiple correlationsbetween performance measures and the biographical variables were obtainedfrom multiple regression analyses using a hierarchical forced entry procedure(Cohen & Cohen, 1983). The experience measures of time in military, time inArmor, time as gunner, and hours on UCOFT were entered into the regressionequation as a functional set of variables. Because the sample included sub-jects with a great amount of experience, experience was hypothesized as hav-ing a larger effect on performance than GT score. Thus, the set ofexperience variables was always entered first into the equation and the GTscore was entered last.

Results

Three groups from Experiment I were combined with a group of subjectsfrom another project to form a two by two experimental design consisting ofthe following groups:

(1) Group 1: No MOPP, normal operational mode.

(2) Group 2: No MOPP, emergency operational mode.

(3) Group 3: MOPP (mask and gloves), normal operational mode.

(4) Group 4: MOPP (mask and gloves), emergency operational mode.

All groups received a pretest consisting of representative exercises from theexperimental conditions. Subjects were not in 14OPP gear for the pretest andperformed under normal operational mode conditions. Four test exercises wereperformed under the experimental conditions according to assigned group.Performance measures of fire time, percent hits, and aiming error were usedin both the pretest and in the experimental conditions.

20

Pretest Performance

A one-way multivariate analysis of varianco (MANOVA) was performed onpretest performance measures to assess group differences in beginning per-formance (see Appendix J for summary data). No significant group differenceswere found on any measure. Therefore, no covariates were used in the analy-ses on experimental test performance. (See Appendix K for intercorrelationsbetween performance measures on the pretest and in the experimental condi-tions.)

Experimental Test Performance

Multivariate analysis of variance (MANOVA) on repeated meatures was per-formed on each measure from the experimental conditions (see Table 9 forsummary data). (See Appendix L for complete summary data on performancemeasures by exercise.) Post hoc comparisons were performed on significanLmain effects and interactions. Planned comparisons were used to examine thedifferences between short and long range targets in the stationary owntank/moving target exercise. Differences between stationary and moving tar-gets were examined in the stationary own tank/long range stationary targetexercise versus the stationary own tank/long range moving target exercise.Finally, the differences between stationary own tank and moving own tank wereexamined in the moving own tank/short range moving target exercise versus thestationary own tank/short range moving target exercise. MANOVA summary ta-bles present complete results of the analyses.

As detailed in the MANOVA tables, the overall main effect for operationalmode was significant for each performance measure, (see Tables 10, 11 and12). Emergency mode significantly degraded fire time, percent hits, andaiming error. There was no significant effect for MOPP gear on any perform-ance measure nor were there any significant interactions between MOPP geare'nd operational mode. The MANOVA tables indicate a significant exercise maineffect for each measure. Planned comparisons between exercises on targetrange indicated a significant effect for long range targets on each perform-ance measure. The comparisons also indicated a significant effect for movingtargets on percent hits and aiming error. A significantly higher fire timeexisted in the stationary own tank exercise and a significantly higher aimingerror in the moving own tank exercise.

The interaction between MOPP and exercise was not significant for anymeasure as illustrated in the MANOVA tables. The interaction letween opera-tional mode and exercise was significant for fire time and aiming error (seeTables 10 and 12). Post hoc comparisons ineicated a significattly slowerfire time in the long range target exercises under emergency mode conditions.No significant differences existed between modes for fire time in the shortrange target exercises. Fire time was significantly longer in the movingtarget exercise under normal mode conditions, however, no diffarence existedbetween the stationary and moving target exercise under emerge-icy mode. Firetime was significantly longer in both the stationary and moving target exer-cisee under emergency mode conditions.

21

Table 9

Summary Data on Experimental Test Performance by Group for Experiment lI

Average Average AverageGroup a . . .Fire Time Percent. Rita Aiming .. Error

1. No MOPPNormal Operational ModeM4 14.20 77.91 1.28 -

SD 1.74 9.30 0.26

2. No MOPPEmergency Operational Mode14 16.09 50.08 1.95

SD 1.15 8.90 0.33

3. Mask and GlovesNormal Operational Mode14 14.43 78.72 1.31

SD 1.97 10.26 0.26

4. Mask and GlovesEmergency Operational Mode14 16.32 53.64 1.93

SD 1.93 10.37 0.35

Not__e. Fire time is the elapsed time (seconds) from full target exposure towhen the gunner fires.Percent hits - (number of hits/number of targets presented) x 100.Aiming error (mils) - square root [((azimuth error) 2 + (elevationerror) 2 ].

an = 12 per group. •

22aw

Table 10

MANOVA Summary Table for Fire Time on Experiment II

Source df F p level

MAIN EFFECTS:

MOPP (Mask and Gloves) 1,44 0.21 usOperational Mode 1,44 14.26 .000Exercise 3,42 70.47 .000

INTERACTION:

MOPP x Operational Mode 1,44 0.00 usMOPP x Exercise 3,42 2.32 usOperational Mode x Exercise 3,42 15.16 .000MOPP x Operational Mode x Exercise 3,42 0.58 ns

PLANNED COMPARISONS:

Target RangeExercise 3 versus Exercise 4 1,44 48.85 .000

Stationary/Moving TargetExercise 2 versus Exercise 4 1,44 2.37 us

Stationary Tank/Moving TankExercise 3 versus Exercise 1 1,44 63.67 .000

POST HOC COMPARISONS:

Operational Mode x Target Range 1,44 8.05 .007Normal Mode x Target Range 1,44 8.62 .005Emergency Mode x Target Range 1,44 48.27 .000Operational Mode x Short Range Target 1,44 0.52 nsOperational Mode x Long Range Target 1,44 8.38 .006

Operational Mode x Stationary/MovingTarget 1.44 10.57 .002

Normal Mode x Stationary/MovingTarget 1,44 11.47 .001

Emergency Mode x Stationary/MovingTarget 1,44 1.46 ns

Operational Mode x Stationary Target 1,44 38.99 .000Operational Mode x Moving Target 1,44 8.38 .006

Operational Mode x Stationary/Moving Tank 1,44 3,67 ns

23

Table 11

... MANOVA Summary Table for Percent Hits on Experiment 1I

Source df F p level

MAIN EFFECTS:

MOPP (Mask and Gloves) 1$44 0.61 usOperational Mode 1,44 88.68 .000Exercise 3t42 76.79 .000

INTERACTION:

MOPP x Operational Mode 1,44 0.24 usMOPP x Exercise 3,42 1.11 usOperational Mode x Exercise 3,42 2.25 usMOPP x Operational Mode x Exercise 3,42 1.39 ns

PLANNED COMPARISONS:

Target RangeExercise 3 versus Exercise 4 1,44 111.51 .000

Stationary/Moving TargetExercise 2 versus Exercise 4 1,44 174.68 .000

stationary Tank/Moving TankExercise 3 versus Exercise 1 1,44 2.85 ns

24

Table 12

MAVOVA Summary Table for Alming Error on Experiment II

Source df F p level

MAIN EFFECTS:

MOPP (Mask and Gloves) 1,44 0.00 nsOperational Mode 1,44 53.42 .000Exercise 3,42 127*59 .000

INTERACTION:

MOPP x Operational Mode 1,44 0.10 usMOPP x Exercise 3,42 2.65 usOperational Mode x Exercise 3,42 19.07 .000MOPP x Operational Mode x Exercise 3,42 0.82 ns

PLANNED COMPARISONS:

Target RangeExercise 3 versus Exercise 4 1,44 41.79 .000

Stationary/Moving TargetExercise 2 versus Exercise 4 1,44 228.26 .000

Stationary Tank/Moving TankExercise 3 versus Exercise 1 1,44 4.84 .033

POST HOC COMPARISONS:

Operational Mode x Target Range 1,44 19.12 .000Normal Mode x Target Range 1,44 2.19 usEmergency Mode x Target Range 1,44 58.72 .000Operational Mode x Short Range Target 1,44 8.54 .005Operational Mode x Long Range Target 1,44 54.92 .000

Operational Mode x Stationary/MovingTarget 1,44 48.83 .000

Normal Mode x Stationary/MovingTarget 1,44 11.47 .001

Emergency Mode x Stationary/MovingTarget 1,44 1.46 ns

Operational Mode x Stationary Target 1,44 1.72 nsOperational Mode x Moving Target 1,44 54.92 .000

Operational Mode x Stationary/Moving Tank 1,44 5.85 .020Normal Mode x Stationary/Moving Tank 1,44 10.67 .002Emergency Mode x Stationary/Moving

Tank 1,44 0.02 nsOperational Mode x Stationary Tank 1,44 8.54 .005Operational Mode x Moving Tank 1,44 0.03 ns

25

As Table 12 Indicates, post hoc comparisons revealed no significant dif-ference for aiming error between the short range and the long range targetexercises In normal mode conditions. However, a significant difference ex-ioted for aiming error between the short range and long range target exer-cises in emergency mode. A significantly higher aiming error occurred wnderemergency mode conditions In the moving target exercise, but no differenceexisted between operational modes in the stationary target exercise. A sig-nificantly lower aiming error occurred In the stationary own tank exerciseversus the moving own tank for those groups under normal operational mode.No difference existed between the exercises for groups In emergency mode.

Pearson Correlations and Multiple Regression

The relationship among performance measures and the biographical varia-bles were examined using Pearson product moment correlations (nee Appendix M1f or biographical summary data). Results again suggest a stronger relation-ship between biographical variables and fire time than with the other per-formance measures. Five of the ten correlations between fire time and thebiographical variables were significant. Only one of the ten correlationsbetween percent hits and the biographical variables was significant. Fur-thermore, only one of the ten correlations with aiming error was significant(see Table 13).

Multiple correlations between the measures of performance on the pretestand in the experimental conditions and the measures of experience and GTscore were examined by multiple regression. A hierarchical forced entryprocedure was used in the multiple regression analyses (Cohen & Cohen, 1983).I The experience measures of time In military, time in Armor, time as gunner,and hours on UCOFT were entered into the regression equation as a functionalset of variables (see Appendix N for intercorrelations between biographicalvariables). Because the sample included subjects with a great amount of

* experience, experience was hypothesized as having a larger effect on perform-ance than the GT score. Thus, the set of experience variables was alwaysentered first into the equation and the GT score was entered last. If theoverall F statistic for the set was significant, the Beta weights of constit-uent variables in the regression equation were examined separately.

The multiple correlation between fire time on the pretest and the func-tional set of experience variables was significant, (R-.63, R2-.40,F(4,42)-6..91; p=.000). The two variables which significantly weighted in theregression equation were: time as gunner (t[42]-2.14; p-.038) and time inmilitary (t[42]=2.05; p-.047). The negative sign of the t value correspond-ing to the negative Beta weight for time as gunner was in the expected direc-tion. However, the Beta weight for time in military was pobitive and not inthe expected direction. The GT score was not significantly weighted In theequation. The multiple correlation between the average fire time in theexperimental conditions and the set of experience variables was also signifi-cant (R-.46, R2=.21, F(4,42)-2.80; p-.038). Two constituent variables weresignificantly weighted in the regression equation: time as gunner(t[42]--2.14; p=.038) and time in military (t[42]-2.26; p-.029). Once again,

26

the Beta weight for time as gunner was negative and In the expected direc-.tion, howeverg the Beta weight for time In military was positive and not Inthe expected direction, The GT anore was not significant when entered intothe equation. There were no significant multiple correlations between per-cant hit* and aiming error and the biographical variables. (Sees Table 14 fordetails of the multiple regression analyses.)

Table 13

Pearson Correlat ions between Performance Measures on the Protest and In theExperimental Conditions and Biographical Variables for Experiment it

Time in Time In Time as CT Hours onMilitary Armor Gunner Score UCOFT

Protest .5048 .4448 -.2727 .1331 -.2093Fire Time ( 48) ( 48) ( 48) ( 47) ( 48)

P-.000 p-.00l p-.030 p-.186 p'm.077

FPretest -.2151 -.2223 .0694 .2028 .2710Percent Hits ( 48) ( 48) ( 48) ( 47) ( 48)

p-.071 p-.064 p-.3 20 P-.086 p-.031

Pretest .1258 .0278 -.1737 -.1732 -.2291Aiming Error ( 48) ( 48) ( 48) C 47) ( 48)

p-.197 p-.426 P-.119 p-.122 P-.059

Experiment .3071 .1470 -.3156 .0370 .0031Fire Time ( 48) ( 48) ( 48) ( 47) ( 48)

P-.017 p-.163 p-.014 p-.403 p-.492

Experiment -.0416 .0977 -.0114 -.0898 .0607Percent Hits ( 48) ( 48) ( 48) ( 47) ( 48)

p-.389 p-.254 p-.469 p-.274 p-.341

Experiment -.0697 -.2520 -.0890 -.0403 -.1392Aiming Error ( 48) ( 48) C 48) ( 7) ( 48)

p-.319 p-.042 p-.274 p-.371 p-.178

27

A11111mii ii

N 44

'.P"40.

f4ON~ ~4~C4 4

a a I

44 . ** C4 C

M UP%

'Ao14

""4 a~t -A

0 0

1 0 41 .4a

1.44 w "4 1 Na =5.4 12 w S

* u- isU a -4 Il 0 r

*4 .54 S S S

0

02

IL14 C4 .

4

%o go m N V4 C4 N

a a"

C4 cc

* ~~W 0t4A ~

.0 'N * D.

o4 od u p4

.0

~r4 r4*~ 4

-o

0 u

14. -. '-I 00j 4

044 -

4) r44 4HEu - o k4-4 1-U4

.0 to 41 CaCAVr- .41 w 4) 41k

290

Ri 4" 40N

I-, Up

en C4 C4

144

Ao .4#4 -8P44 r6. 0 00

0~ '0 a a a

0 0 1 wC 41.4 a Q44

300

LammobMIM i b fth

Discussion

...... ... .The results of this experiment revealed a significant overall effect foremergenvy mode on gunner performance while no effects for IIOPP gear wereindicated on any performance measure. The significant effect for ?fOPP gearon aiming error found in Experiment I was not replicated in this experiment.Emergency mode significantly degraded all three performance measures: firetime, percent hits, and aiming error. Furthermore, no significant interactionbetween ?IOPP gear and operational mode existed. The nonsignificant interac-tion indicates that MOPP gear did not additionally affect performance whenunder emergency operational conditions. Emnergency mode remained the onlysignificant effect on gunner performance.

The results of this experiment also suggest that the degrading effects ofemergency mode may depend upon situational characteristics. For example,emergency mode only affected fire time in long r 'ange target engagements.Aiming error was affected by emergency mode in the moving target exercisewith no significant effect in the stationary target exercise. Another resultindicated significant effects for emergency mode on aiming error in the sta-tionary own tank/moving target exercise and-no significant effect for emer-gency mode in the moving own tank/moving target exercise. These resultssupport the conclusions of Experiment I.. Performance decrements cannot beassumed for all target engagements.

The multiple correlations once again suggested a relationship betweenfire time and experience, while the accuracy measures of percent hits andaiming error were not significantly correlated with experience. The positivedirection of the Beta weight for time in military, however, suggests a to-tally different relationship with fire time. The positive direction indi-cates that as time in the military increases, fire time increases, hence, theslower the gunner's speed at firing a- round of ammunition. This is contra-dictory to the negative relationship between fire time and time as gunner.

GENERAL DISCUSSION

This research was designed to examine and identify possible deficienciesin Armor crewmen performance while wearing NBC equipmernt under normal and de-graded operational conditions. Previous research has revealed large perform-ance decrements associated with NBC protective equipment (Muza, 1986). Theperformance degradation largely corresponds to visual limitations due to therestrictions imposed by the mask and loss of manual dexterity when wearingthe rubber gloves. Research indicates that measures of speed may be affectedwhile accuracy is not (Fine & Kobrick, 1986; Rauch et &l., 1986). Existingresearch on tank crewmen under NBC conditions has focused on performancedecrements due to the physiological effects of fatigue and heat stress fromsustained operations in complete MOPP gear corresponding to the MOPP4 level(Carr, et al., 1980). Because tank gunnery relies heavily on visual capabil-ity and manual dexterity, it is imperative to investigate and identify theeffects of MOPP gear on psychomotor responses in gunnery performance whilecontrolling for fatigue and heat stress. Future battlefield conditions canresult in malfunctions in the tank's normal operational mode, therefore,

31

examining the effects of MOPP gear in normal and degraded operational condi-tions is warranted.

Results of this research did not support previous indications of largeperformance decrements associated with NBC equipment. The only significanteffect for MOPP gear was for aiming error in Experiment I. However, thisfinding was not replicated in Experiment II. In Experiment II, the onlysignificant result was the degrading effect of emergency operational. mode onall performance measures. There vere also no significant interactions be-tween MOPP gear and operational mode. This result suggests that MOPP geardid not affect nor further degrade performance under emergency operationalconditions as might be expected. As Muza (1986) suggested, the effectivenessofweapon or-surveillance systems should decline because the protective maskhinders the alignment of the eye with the optical sight. However, alignmentof the eye with a weapon system's optical sight constitutes a narrow field ofview. Hence, "... performance of certain tasks may be enhanced by CB[chemical, biological] mask wear . . . the narrower field of view may elimi-nate distractions and help the soldier concentrate on his task" (Muz, 1986,p. 18). The results of this research support Muza's suggestions, but shouldbe further investigated for more conclusive evidence.

A factor which may have affected the results for MOPP gear is the experi-ence level of the subjects. They were highly qualified and experienced Armorpersonnel. They were instructors from the M1 NETT team, trained to instructMI gunnery and conduct of fire procedures. On the average, the subjects hadmore than 12 years of time in the military and more than two years overalltime as gunner. These qualifications alone suggest experience as a possiblefactor in their successful performance. The multiple correlation betweenfire time and the experience variables, particularly time as gunner, wassignificant. The negative direction of the correlatioa for time as gunnerwas expected. Gunners should become faster at firing a round of ammunitionas their experience as a gunner increases. However, the correlation betweenfire time and time in rilitary, while significant, was positive. The rela-tionship may be due to the slowing of response times with age which has beenfound in numerous research concerning aging effects on motor behavior(Hodgkins, 1962; Noble, 1978; Schmidt, 1982). "One simple generalization isthat past the age of about 25 years, a progressive decline occurs in justabout every measurable aspect of motor behavior . . . one of the most consis-tent findings in the aging literature is that people become slower with age"(Schmidt, 1982, pp. 426-427). Therefore, the positive direction of the cor-relation may simply support previous reaction time research.

The positive results for MOPP gear on gunner performance in this sampleof experienced armor personnel support the conclusion of Fine and Kobrick(1986). Fine and Kobrick (1986) examined the sustained performance oftrained personnel in complete MOPP gear on military tasks routinely performedduring an NBC attack. The subjects were exposed to heat stress and a controlcondition with cooler temperatures and lower humidity. From their results,Fine and Kobrick concluded that ". . . very well trained soldiers in MOPP4,sedentary, and performing cognitive tasks in relatively moderate heat andhumidity conditions, on the whole will probably function effectively for upto three hours . . ." (,age 90). Their results indicated a dramatic drop insubjects' performance after four to five hours of heat exposure while in the

32

MOPP4 level of NBC protection. However, heat stress appeared to be a moredecisive factor in the performance decrement than the MOPP gear alone. Afterseven hours, the overall performance of subjects in MOPP4 when exposed tocooler temperatures was not significantly different from their performance inthe control condition when not wearing MOPP gear. Therefore, the results ofFine and Kobrick and this research project suggest that MOPP gear alone doesnot necessarily result in large performance deficiencies. It appears to bethe combination of wearing MOPP gear during sustained operations and exposureto hot and humid conditions that impairs performance.

Training in MOPP gear under NBC conditions is highly stressed in today'sArmy. Presently, "our adversary is better equipped and trained to operateunder NBC conditions than we are-and could capitalize upon such an advantagein military operations" (Moffett, 1981, p. 38). The solution to this problemlies in training for the future battlefield during peacetime. In order tominimize the threat of NBC weapons and reduce the performance degradationassociated with protective equipment, soldiers must periodically train inMOPP gear under NBC conditions. Success on the modern battlefield will re-quire the soldier to conduct operations and accomplish his mission in protec-tive equipment. "Unless the unit is well-trained and conditioned in NBCprotective operations, the loss of operational effectiveness ... will have

an adverse effect on mission accomplishment" (FM 17-17, p. E-17). Welltrained and thus well-prepared soldiers will suffer less overall stress whenusing MOPP gear in combat. Therefore, training under NBC conditions in MOPPgear is the key to reducing casualties in actual combat and increasing opera-tional effectiveness. "Training is the cornerstone of success . . . on theday of battle, soldiers and units will fight as well or as poorly as theywere trained before battle" (FM 100-5, P. 1-4).

Considering the extensive amount of time in the military and in Armor,and overall time as gunner for this sample of subjects, it is not unrealisticto assume that the majority have received a fair amount of NBC training.Their successful performance while wearing MOPP gear may be a reflection oftheir previous training. A number of subjects informed the researcher thatthey had received a large amount of training in MOPP gear. However, no ob-jective data were collected for confirmation. Therefore, the validity ofthis hypothesis awaits determination by research using less experienced Armorcrewmen and obtaining quantitative data on NBC training.

Extensive training under degraded operational conditions appears manda-tory to offset the performance deficiencies revealed in this research. Formission accomplishment, soldiers should be trained under all potential condi-tions faced on the modern battlefield including limitations in operationalreadiness. The UCOFT training matrix includes exercises specifically de-signed for training under various degraded operational conditions. There-fore, UCOFT training should be considered a critcal component in trainingcountermeasures designed to alleviate the level of degradation associatedwith future battlefield conditions. As Fine and Kobrick (1986) stressed, theconsequence of overlearning routine tasks in realistic military scenarios iswell-trained soldiers.

In summary, the results of this research indicated no substantial defi-ciencies in gunnery performance associated with the protective mask and

33

11:110 51 , '' ý11 IIII I

Um-A5MND WO

gloves in this sample of subjects. At the same time, deficiencies existedwhen gunners were required to perform under less than optimal operationalconditions. A number of factors could have affected the positive results forHOPP Sear which require. further investigation. The performance degradationassociated with emergency operational mode strongly suggests the developmentof intensive training countermeasures to moderate the impact of possiblebattlefieJA threats on the gunner's proficiency. P.ealistic training inpeacetime results in real soldier preparedness.

34

.. .. . . ... . .

REFERENCES

Barnes, J. A., Bruno, R. S., Hianlon, W. E., Harrah, D. M., Hickey, C. A.,Merkey, R. P., Randall, R. B., & Shoemaker, C. M. (1983). XM30Engineering design test -. Covernment (Human Factors). (Tece TnalMemorandum 5-83). AberdeeL Proving Ground, MD: U.S. Army HumanEngineering Laboratory.

Bensel, C. K. (1980). A human factors evaluation of two types of rubber CBprotective gloves. (Technical Report Natick/TR-80/005. Natick, MA: U.S.Army Natick Research and Development Command.

Carr, J. L., Kershner, R. L., Corona, B. M., & Jackson, S. E. (1980). Theeffects of CB clothing and equipment on U.S. Army soldier performance: Acritical assessment of performance testing. (Technical Memorandum25-80). Aberdeen Proving Ground, MD: U.S. Army Human EngineeringLaboratory.

Cohen, J. & Cohen, P. (1983). Applied multiple regression/correlationanalysis for the behavioral sciences. Hillsdale, NJ: Lawrence ErlbaumAssoc.

Cox, T. J. & Jeffers, A. R. (1981). Ground crew chemical defense equipmentperformance task-time degradation test. (Technical Report 81-5003).Dayton, OH: U.S. Air Force Aeronautical Systems Division.

Fine, B. J. & Kobrick, J. L. (1986). Assessment of the effects of heat andNBC clothing on performance of critical military tasks. (TechnicalReport T11-85). Natick, MA: U.S. Army Research Institute ofEnvironmental Medicine.

Hodgkins, J. (1962). Influence of age on the speed of reaction and movementin females. Journal of Gerontology, 17, 385-389.

Janson, J. P. & Jepson, G. W. (1982). Dexterity degradation study phase II.Wright Patterson Air Force Base, OH: Air Force Aerospace MedicalResearch Laboratory.

Johnson, R. F. & Sleeper, L. A. (1986). Effects of chemical protectivehandwear and headgear on manual dexterity. In Proceedings of the HumanFactors Society, Santa Monica, CA, 994-997.

King, J. M. & Frelin, A. J. (1982). Patient care in a chemical environment.substudy: Effects of protective clothing on the performance of basicmedical tasks. (Final Report 83-0011). Fort Sam Houston, TX: U.S. ArmyHealth Services Command.

Kobrick, J. L. & Sleeper, L. A. (1986). Effect of wearing chemicalprotective clothing in the heat on signal detection over the visualfield. Aviat. Space Environ. Med., 57, 144-148.

35

Moffett, D. L. (1981). NBC and the Armor crewman. Armor, September-October,38-41.

Montague, W. E., Baldwin, R. D., & McClure, A. H. (1959). The effects ofwearing the CBR mask upon the performance of selected individual combatskills. (Technical Report 57). Washington, D.C.,: HumRRO.

Muza, S. R. (1986). A review of biomedical aspects of CB masks and theirrelationship to military performance. (Technical Report (T1-87).Natick, MA: U.S. Army Research Institute of Environmental Medicine.

Noble, C. E. (1978). Age, race, and sex in the learning and performance ofpsychomotor skills. In R.T. Osborne, C.E. Noble, and N. Weyl (Eds.),Human variation: The biopsychology of age, race, and sex. New York:Academic Press.

Rakaczky, J. A. (1981). The effects of chemical protective clothing andequipment on combat efficiency. (Technical Report 313). AberdeenProving Ground, MD: U.S. Army Material Systems Analysis Activity.

Ramirez, T. L., Shew, R. L., Felt, J. E., & Rayle, M. E. (1986). A methodfor determining task time increase caused by the individual protectiveensemble. (AAAMRL-TR-86-036). Wright Patterson Air Force Base, OH: AirForce Aerospace Medical Research Laboratory.

Rauch, T. M., Witt, C., Banderet, L. E., Tauson, R., & Golden, M. (1986).The effects of wearining chemical protective clothing on cognitiveproblem solving. (Technical Report T18-86). Natick, MA: U.S. ArmyResearch Institute of Environmental Medicine.

Schmidt, R. A. (1982). Motor control and learning: A behavioral emphasis.Champaign, IL: Human Kinetics Publisher.

Smith, E. P. & Graham, S. E. (in preparation). Validation of psychomotor andperceptual predictors of Armor officer M-1 gunnery performance.Alexandria, VA: U.S. Army Research Institute for the Behavioral and SocialSciences.

U.S. Department of the Army. (1984). NBC Operations. Field Manual(FM 3-100). Washington, D.C.: Headquarters, Department of the Army.

U.S. Department of the Army. (1985). NBC Protection. Field Manual (FM 3-4).Washington, D.C.: Headquarters, Department of the Army.

U.S. Department of the Army. (1982). Operations. Field Manual (FM 100-5).Washington, D.C.: Headquarters, Department of the Army.

U.S. Army Armor School (1983). The Division 86 Tank Battalion/Task Force.Field Manual (FM 17-17). Washington, D.C.: Headquarters, Department ofthe Army.

36

U.S. Army Aviation Center (1985). Army Aviation in a NBC Environment. FieldManual (FM 1-102). Washington, D.C.: Headquarters, Department of theArmy.

Wagner, R. L. & Gold, T. S. (1982). Why we can't avoid developing chemicalweapons. Defense 82, July,3-11.

Wiley, R. W., Behar, I., Chiow, W. C., & Holly, F. F. (1977). Visual andoptical analyses of XM-29 and M-24 protective masks. (Technical Report77-14). Fort Rucker, AL: Army Aeromedical Research Laboratory.

Witmer, R. G. (in preparation). Effects of degraded mode gunnery procedureson the performance of M1 tank gunners. Alexandria, VA: U.S. Army ResearchInstitute for the Behavioral and Social Sciences.

37

APPENDIX A

LISTING OF EXERCISES - EXPERIMENT I

PRETEST EXERCISES

Conditions - Fully Operational TankNormal Operational ModeDay Unlimited VisibilitySingle Targets

UCOFT ExerciseNumber

31111(0) - Stationary Own Tank/Short Range Stationary Target32111(0) - Stationary Own Tank/Long Range StationaryTarget31311(0) - Stationary Own Tank/Short Range Moving Target32311(0) - Stationary Own Tank!Long Range Moving Target31411(1) - Moving Own Tank/Short Range Stationary Target32411(0) - Moving Own Tank/Long Range Stationary Target31511(0) - Moving Own Tank/Short Range Moving Target32511(1) - Moving Own Tank/Long Range Moving Target

EXPERIMENTAL CONDITIONS

Conditions -

Groups 1, 2, 3: Fully Operational TankNormal Operational ModeDay Unlimited VisibilitySingle Targets

Group 4: Computer, Laser Range Finder, Stabilization System,and Gunner Primary Sight - Nonfunctional

Emergency Operational ModeDay Unlimited Vis-bilitySingle Targets

UCOFT ExerciseNumber

32131(0) - Statinnary Own Tank/Long Range Stationary Target31431]0) Mo-. Own Tank/Short Range Stationary Target3243i(W-) &32431(1) - Moving Own Tank/Long Range Stationary Target31331(0) - Stationary Own Tank/Short Range Moving Target32331(0) - Stationary Own Tank/Long Range Moving Target31531(0) - Movi:' 'ý'n Tank/Short Range Moving Target

A-I

APPENDIX B

INSTRUCTIONS TO SUBJECTS

The purpose of this research project is to determine how gunnery perform-ance is affected by degraded engagement conditions and MOPP gear. You willbe required to engage a series of computer-generated targets on the UCOFTunder normal or degradecý conditions. Prior to engaging these targets youwill perform warm-up exercises to introduce you to the UCOFT. Questionsabout the UCOFT or about experimental procedures should be asked during thewarm-up exercises. After the warm-up you will be given a 10-minute break andwill complete a biographical questionnaire. You will then return to theUCOFT to perform the test exercises. Those of you who will be in MOPP willthen put on the appropriste gear. For the test, you will perform two sets ofthree exercises each, separated by a 10-minute break. No assistance will beprovided by the UCOFT Instructor/Operator or by the TC on the test.

The UCOFT presents both offensive and defensive scenarios. In the offen-sive scenarios you will usually shoot with your tank on the move. In thedefensive engagements, your tank moves up from behind a berm, stops, and thenfires while stationary. Your instructor/operator will inform you prior to anexercise whether you are in an offensive or defensive posture.

The UCOFT is capable of simulating both normal and degraded conditions.The instructor/operator will inform you of the system degradation if it ex-ists for each exercise prior to the start of the exercise. The instructor/operator will also instruct you on the switch settings for the exercise andindicate the procedures to be used in that exercise.

As each target is presented, your TC will lay the gun on a landmark nearthe target and issue a fire command. Because fire commands will vary fromone engagement to the next, you must listen very carefully to each fire com-mand. When you hear the TC command, you should search for the target withyour GPS or GAS, and announce IDENTIFIED upon acquiring the target. When youdetect a target with the GPS, you shou'2 switch the GPS to IOX, lay on thetarget, track it if it is moving, and fire upon the TC's command announcingON THE WAY when you fire. Your speed and accuracy of engaging targets isimportant and will be measured for each engagement. When you hit a target, awhite flash occurs at the target, its normal motion ceases, and it assumes akilled posture. If you miss a target dirt will be kicked up. If you fail tokill the target in the allotted time (18 to 24 seconds), all sights will goblack and controls will be inoperative for five seconds to indicate that youwere killed. The gun select switch will return to the TRIGGER SAFE positionwhen you are killed and must be reset to the MAIN GUN position prior to fir-ing the next engagement.

B-I

80 C4 cc% M0

.2~- mflt SA? .-qq44

enD N

.4V1T (4%a 0~ cc %0 001(n en en *T S *

H 00 w-4

on0 .4-

410

4) E-

-E 4

3.'> 4J)- oi4

H 4 14 80 0

04 0 a

04 a

04 4 .4q .4

4. 0

to 0 A 4.

U~s 0 wO-1

APPENDIX D

BIOGRAPHICAL QUESTIONNAIRE

Subject I ..... _ Group __Date

1. Age years

2. Grade E-

3. Education level. Circle one.

a. less than 12 years b. GED c. high school graduate

d. technical school e. some college __ years

f. college graduate g. other (describe)

4. General Technical (GT) Score Social Security No.

5. Total time in service - years - mos.

6. How long have you been in Armor? years MoS.

7. Present crew position Time in position mos.

Present vehicle

8. Time spent as a gunner Mos. Time as MI gunner Mos.

Time as M60A3 gunner mos. Other gunnery time Vehicle

9. When was your last training/sustainment gunnery practice? moS.(Exclude COFT)

10. on how many separate occasions have you fired the COFT? Hours on

COFT

D-1

to P4a C4 n I* N f

V4f~ tt4

"r44 41

1--41

141 4 4

pq 10

V-44

14 4 4C

14

44 0

'A

in 14.

4C. 440 0 1 41 A $ S *

1441 - -4 ,Ve a. ~ 4 441

Ow 014 "~

a. t- 41 Vo Va 9

r-I4 431'4. 4 0

-4 ii as UoV1l-4 v4

41 w a 04 0--MI I. a -4 0) ab.

M. W1 r. W1 W44'a0~ 0M01a21

00l 0 CU41*1

'a ".4 4~~E14 414*43

14 43 a ~ 'a 41 41 414 0

APPENDIX F

INTERCOPULATIONS AMONG PERPMOANCR MEASURLS - EXPERIMENT I

Table F. 1

Pearson Intercorrelat ions on Performance Measures oan the Pretest and In theExperimental Conditions for Experiment I

Pretest Pretest Protest Experiment ExperimentFire Time Percent Hits Aiming Error Fire Time Percent Hits

Pretest -. 3044Percent Hits ( 48)

p-.018

Pretest .1704 -. 8247Aiming Error ( 48) ( 48)

p-.123 p-.000

Experiment .6153 -. 1087 -. 0278Fire Time ( 48) ( 48) ( 48)

p-.000 p-.231 p-n.426

Experiment -. 0744 .2528 -. 2529 -. 4625Percent Hits ( 48) ( 48) ( 48) ( 48)

p-.308 p-.041 p-.041 p-.000

Experiment -. 1307 -. 1068 .1835 .2488 -. 6786Aiming Error ( 48) ( 48) ( 48) ( 48) ( 48)

p-.188 pr. 235 pmo 106 p-.044 p-. 000

F-I

bUN inN N P

1.4O.

.P44-

taa

-4v 00 4n0 oC ff1

baa

bC3

ta "4

411

be -I

AaA

044

G-1

0ih

41

$4 14 4i ý% L ou* nI

> 44

41 0)04 .0 %C m 0--_ m m%

boba1

.14 0i~ 0V4 U j >14 C44 uu

E-4 1*0 4) 0' f41 I%-144

1.4 c40)-~CC

14- Ai141

0

co ca-P 0

> 0

0 414 r44 4 )$

0 4 1-4

00 004 r-1 14 1414 %0d %c J 0go.. .

0b 4 1 4 -

$i- 0

.44 C4 G2C

Ln C 0 1 %,G WLN ( %0h

00164 U"

f41

0)b 4 0 cO

$4 o, I04'0iU

0.

"-4

0 400 0

bo

00 4)

0N b01a . .* **

V4~~~ ~ .".0 0'4.4coc C -4 q)

u 0 0 00

00

C4.

u~0)ob

.0 b 44.0 0~140

44)

1.4N C% ý%0-40 C.e4 c.&A 0.4)$40 :P% ca CO 0% . .n ..

0 go 0;,@ o40 0 bo CO V0-4 0'. r- 0-4 03

4 -4 009O ~ r 0.0 .

"4.J1.44

0 ca c.c4c-A. 0)" ) $

04 00 " 00 c0 cco

*0 4 64 4.) 0

A0 0 00 Vo 0 0)1.4 1. ý-1. r4O 1 0

C1. p 0 0o 00 5Ow 04 a01.

E-5ol 4 C4.40440..

G-3

N %-9. -4 % mI ~UF I00 C'0 LM

Go%

Hx

1.4400

0 441

0) 40oo b

Un 0

dL 0C4L 00 r n 4

o 0 ob f 4 0- n$0 C: 4

0 000

004

0 40444 0

0 4.1

-r4.0

> 0

0 0 to00 tC

1 to 0cc 0 cc0 d94 *aW81410 0 0) 41 4)

0 0 V0 3Q4

r40 41 t- 0o 4)4 40 0 co 0~ f0 j) C

g 04 0to 040A w .4 940.94 el 4.5 C

co 0) 004 c.ro I-

0.v- Ov- 0 0G-4

L 04 O vI00

"$41-4 sN V 0 O M%

C00

t~ic(a4 r- N04c 0 t"4

0~~~ "44 0.4 C4 ~ q4

4) 0 C4"0 4 to)g

Li 0

0 0iNd4 $4 4)

E-4 cc in In n0 0,-4 O N 00 N 4)a)N C4b CA . 1 oC 0N r-. V-4 I,. C4 H

4a NbC3 41 .

0 0 +

0)0

W Li0$4)

$4)$0

.0W bo4 co

OO clc ; ;4

0

1-r4 r00

Ai 4to>- 0~

-4)cnj

00 00c 03Io 4 r4 "4 4)

*14 v~ 41 > %$4. go 0 0 & 4)

H4 H- 004) 0 04

00 0 00 1JW 0W 0l 4) 14 L

.1-4 .1- r-4 9~ 0 *

0n 03 0 00 00. "I Li 04

0c E) 4) AD)jD 4L

00 1.0

04) 4) 4)004 04 c0 zOH' 04

G-5U M~

rO% C~~~~C'4 '' -

41.

4.4 1

cc14

-41- 0

o >-.

ca ON10

00

011-44

E- 01000

5V4 H401 01

-A-

.0

01 100

"45.

0 05A.4c g

014 $44 O .4 4 4 041 to10 CA.0cao0 4 4

93 -$ $ H$480r- 4.) 00 - 50 -4) ' r-' 0 '- a)c' w

o 01 "40 '-9 4. r-40 '- "-

C; 0101 931

-C4

go 0

go o$4 01 01 C;

G-6S

4.1 0 H 1MI

APPENDIX H

INTERCORRELATIONS AMONG BIOGRAPHICAL VARIABLES - EXPERIMENT I

Table H.1

Pearson Intercorrelations on Biographical Variables for Experiment I

Time in Time in Time as GTMilitary Armor Gunner Score

Time in .8260Armor ( 48)

p-.000

Time as -. 2008 -. 0809Gunner ( 48) ( 48)

pm.086 p-.292

GT -. 0286 -. 0248 .0134Score ( 47) ( 47) ( 47)

p-. 4 24 p-.434 p-. 4 6 4

Hours on -. 1102 -. 0142 -. 0083 -. 1309UCOFT ( 48) ( 48) ( 48) ( 47)

p=.228 p-.462 p-.478 p-.190

H-I

APPENDIX I

LISTING OF EXERCISES - EXPERIMENT II

PRETEST EXERCISES

Conditions - Fully Operational TankNormal Operational ModeDay Unlimited VisibilitySingle Targets

UCOFT ExerciseNumber

31111(0) & 31211(0) - Stationary Own Tank/Short Range Stationary Target32111(0) & 32211(0) - Stationary Own Tank/Long Range Stationary Target31311(0) & 31331(l) - Stationary Own Tank/Short Range Moving Target32311(0) & 32331(1) - Stationary Own Tank/Long Range Moving Target31511(0) & 31531(0) - Moving Own Tank/Short Range Moving Target32511(1) & 32531(0) - Moving Own Tank/Long Range Moving Target

EXPERIMENTAL CONDITIONS

Conditions -

Groups 1, 3: Fully Operational TankNormal Operational ModeDay Unlimited VisibilitySingle Targets

Groups 2, 4: Computer, Laser Range Finder, Stabilization System,and Gunner Primary Sight - Nonfunctional

Emergency Operational ModeDay Unlimited VisibilitySingle Targets

UCOFT ExerciseNumber

31531(0) & 31511(0) - Moving Own Tank/Short Range Moving Target32131(0) & 32111(0) - Stationary Own Tank/Long Range Stationary Target31331(0) & 31311(0) - Stationary Own Tank/Short Range Moving Target32331(0) & 32311(0) - Stationary Own Tank/Long Range Moving Target

I-1

1.42

%a1 C4 . r-4 %0 P-NV40 "40 . -

4j 1

131

0 %.

04 41 1

ba"V 64 w

4-4 0 r 4-.C eq eqW C % ALnAto E40 -4 r4.- C4 %0 4OI 0% 0 44J 4

1- . 41 uOC 64 C40- to 4 0.-

0.

41 94.1 a41.

044 0

0 414 a A"4 E

41 414 P4

134141 U W4

0 'o1 * 1$$41-r4I $4 '.a -4j01a

64 41 "> 1 4 41to ad 0to 0 411 , .0

34 $44 r-4. 41.

4) 4)C & r4A

4.311 "41

V-4 0) U) 4) wa 0 o 0q4 "

21~4. ='4409 o41 4M. v4

04 0 a.0 a04 4$00

E -4 z04C4cn-tZ

41 41 .54 4J 1 " 1.M

APPENDIX K

INTERCORRELATIONS AMONG PERFORMANCE MEASURES - EXPERIMENT II

Table K. 1

Pearson Intercorrelations on Performance Measures on the Pretest and in theExperimental Conditions for Experiment II

Pretest Pretest Pretest Experiment ExperimentFire Time Percent Hits Aiming Error Fire Time Percent Hits

Pretest -. 3830Percent Hits ( 48)

p-.004

Pretest .1947 -. 7780Aiming Error ( 48) ( 48)

p-.092 p-.000

Experiment .4099 -. 0414 -. 0468Fire Time ( 48) ( 48) ( 48)

p-.002 p-.390 p-.376

Experiment -. 0119 -. 0195 .0113 -. 6054Percent Hits ( 48) ( 48) ( 48) ( 48)

p-.468 p-.448 p-.470 p-.000

Experiment -. 1234 -. 0380 -. 0266 .4183 -. 7721Aiming Error ( 48) ( 48) ( 48) ( 48) ( 48)

p-.202 p-.399 p-. 4 29 p-.002 p-.000

K-I

19 10t4'.4- U o-,4q

41.

34V4

'.4

1-4 ba

A.

-P4 1- W jc14 fý94G C4O Ao

0..

'.4 4 bGa-

W -4 41-4 S4c 1 N"

C4 CO 5flU 8 00

ta 04 00r4GoC'.4 14.

u ) 14 0

Z W.4 44

'-4$4. ra

V-4 0

0 > 4

44 O

-r4 -AO0~O004W- W- * ** **

) V4U 14IatAi>C

00

R io a1 ba asU V1.w I4 M $4 I 4

$4' 0 41 0j0

0 0 AId 0 44

. .i n I . n P.km-ae r00 C4 Ul% ;

W% -- - - 1%4 "P f

1-4-414.

Or. t't4'401%Int 0' 'C- 0%0%in (ne

*n M% e N r%M

*~~" 0f(4M%

4))

0Ga

ad.D40 01 00fl M

@00$ r%g) -4 00 4 tO ~~ ~~ r4 0~ '1 MAMI

0, :0 05 "4bAX: .4.0 ca

0. A414

444

S1 00-40 A-0

0~ 0-I41 14 1

0 u40 q)

W 04J 0 u $

4 0 U)

f 0

40 ) 14.E- 4 C40 5.4 t

00-

t Wiflm

01,

"Ue

$4 a

toAj.1

04 H4

14 a

14 4J 0

0 0

'a 03

* 0140

"0 140 +

c'4 014 W

bo tk

14 "4 $

o 'Ib r4"4 C

X. MO4 114 114

P014

0 ca

L-3 '

N~ C4C4 P4C 4C .

N 4. fn(

1554

".4FAc %a

0; C4V cn.- 04 00% w mf0 * 55 5 5 U 5

"" -4 4

I-.4

r4~ S5 * 56 54

*1*0

4J 14

S.'.

Ct '-4 -

I9 e oo%%(b 4% (

A 0

M- 1

APP3INWX N

I NTTIONS AMN0G BIOGRAPHICAL VAWIABIU - RIPIII NT I1

Table N.I

Pearson Intereorralations on Biographical Variables for kUprinent II

Tim in Time in Time an GTMilitary Armor Gunner Score

T im m In .7 8 6 2- - -Armor ( 48)

p-.000

Time as .0076 .0338Gunner ( 48) ( 48)

pa..480 p-.410

CT -. 1805 -. 0886 -. 1360$cre( 47) ( 47) ( 47)p-.112 p-.277 pr.181

Hours oan .0328 .1302 .0272 -. 1167"UCO "" ( 48) ( 48) ( 48) ( 47)

p-.412 p-.189 p-.427 p-.217

14-1