Aircraft pilots are faced with numerous challenges during flight that if not carried out correctly, the pilot could be overcome causing the loss of control and effectively crashing his/her aircraft. Kristen L. Casto and John G. Casali completed a study with 20 Military Helicopter Pilots within the U.S Army Aeronautical Research Laboratory using a full motion helicopter flight simulator. Their study was to determine the effects of hearing loss, aviation headset type, flight workload, communication signal quality while monitoring the pilots performance during simulated flight.
Primarily this study is about pilot performance level and the effects on pilot performance level when Human Factor variables are introduced. Three steps of increasing workload min, medium, high along with three different types of headsets were introduced. Included is the pilots hearing ability and communications message quality is also assessed. Refer to Table. 1 - Headset Types. Knowing this information we can highlight the Independent variables, and the Dependent variable. We can identify the Hypothesis and the Null Hypothesis, and as an extra, we can infer a prediction that will either support or will reject the Null Hypothesis. However Statistical significance will also identify which way to support these theory.
Identification of Hypothesis
- H0 =
- The introduction of three different head sets and three increasing levels of workload, including hearing ability, and communications message quality, will not induce a change in the pilots performance level.
- H1 =
- The introduction of three different head sets and three increasing levels of workload, including hearing ability, and communications message quality, will induce a change in the pilots performance level.
The analysis of the collected research data within the original case study, tests H1. To help clarify I have identified the dependent and the independent variables. Thus if a change is observed in the dependent variable this should be caused by a change in the independent variable or other influencing errors. Please refer to Figure. 1 for identification of variables. Thus I infer that the introduction of the stated independent variables into the research experiment will provide high probability that H0 will be rejected. This could be supported by Statistical significance when completing the analysis of the research data collected during the experiment.
Figure. 1 - Identification of Variables
|Figure. 1 - Identification of Variables|
|Pilots Level of Performance||Type of Headset|
|Pilot Hearing Ability|
|Communications Message Quality|
Table. 1 - Headset Types
|Table. 1 - Headset Types|
|Bose Aviation X ANR Headset||B-ANR (Active Noise Reduction)|
|David Clark Model H10-66 (Passive)||DC-CEP (Communications Earplug)|
|David Clark Model H10-76 (Passive)||DC-FOAM (Foam Earplug)|
To improve accuracy within the sample for the identified population, the 20 Military Helicopter pilots were sorted into two groups based on their hearing sensitivity. 1. Non-Hearing Waived and 2. Hearing Waived group. The pilots within the Non-Hearing Waived group have hearing that is considered better than the Army's Flight Standard requirements. While the pilots within the Hearing Waived group, have hearing that is considered equal too or worse than the Army's Flight Standard requirements. Refer to Table. 2 - Participant Demographics which illustrates the Age and Flight Experience of the 20 pilots.
Table. 2 - Participant Demographics
|Table. 2 - Participant Demographics|
|Variable||Non-Hearing Waived||Hearing Waived||Groups Combined|
|Experience Flight Hours Range||75 - 12,000||1,100 - 11,000||75 - 12,000|
|Experience Flight Hours Median||200||4,350||1,900|
|Experience Flight Hours Average||1,678||4,770||3,228|
During the grouping of the pilots the experimenter discovered an interesting unintended variable. Within the figures of Table. 2 the Hearing Waived group is shown to have more flight experience than the Non-Hearing Waived group. Both Kristen L. Casto and John G. Casali identified this data as a confounding variable however they present this data within their experiment as a realistic representation of the Army's Pilot Population.
Rather than completing the experiment while flying operational helicopters and possibly endangering flight crews, the 20 Helicopter pilots were tested on a Black Hawk Helicopter flight simulator which is a fully hydraulic actuated motion based system. All the controls and the environmental systems within the flight simulator are an accurate representation of what is actually fitted and used to control a fully operational Black Hawk Helicopter. As this study is primarily about testing pilots hearing and speech communications an ambient noise within the flight simulator were introduced to match that of a normal operational flight. The introduced ambient noise level was set at 95dBa.
|Typical NUH-60 Flight Simulator|
DESIGN OF THE EXPERIMENT:
Three different types of headsets were chosen within the experiment to cover the most widely used configurations. ANR(Active Noise Reduction), (CEP) Communications Earplug and Passive Headset with foam earplugs worn underneath the ear cup. It has been recognized that increased mental workload can have an influence on pilot performance during critical times when piloting aircraft, therefore each pilot was subject to three levels of communication intelligibility. The pilots perceptual workload was also influenced by three levels of visibility during flight. Finally the pilots Psychomotor was controlled by introducing increased levels of required flight maneuvers during the experiment. As an example of a low-workload flight segment, a radio comms message required one flight task. Workloads are increased using longer comms messages with mandatory pilot read-backs and increased number of flight tasks. Each pilot flew three flights which was one flight with each different headset.
Table. 3 - Sample Radio Comms & Sample Pilot Read-back Comms
|Table. 3 - Sample Radio Comms Message with Required Flight Task||Table. 3 - Sample Pilot Read-back Messages|
|Turn right heading 270 Degrees||Altimeter 29.92, winds 320@12|
|Turn left heading 180 Degrees||Runway 06 in use, winds 060@5|
|Turn right heading 290 Degrees; climb and maintain 2,500ft||Altimeter 29.90, cleared to land runway 06|
|Turn right heading 350 Degrees; climb and maintain 3,000ft||New altimeter 29.93, winds 040@20|
|Turn right heading 270 Degrees; climb and maintain 2,500ft; descend to 100kn||Winds aloft 290@10, information Charlie is current|
|Turn left heading 190 Degrees; descend and maintain 2,500ft; decelerate to 90kn||Runway 36 in use, winds 33@12|
Within this experiment, Casto and Casali tested their data for Statistical Significance using an Alpha level of 0.05. The reason for this test, is to identify the differences in the means which can be changed and influenced by the experimental variables. During the analysis of the flight control performance data the results showed little or no significance between the three different headset types and the groups of pilots. Although when the group of pilots where averaged together, they showed increases in deviation from Heading, Altitude and Air Speed with all p values at less than 0.003 as the experimental workload was increased. To clarify; with alpha level set at 0.05 and the above stated p values well below 0.003 the hypothesis of Heading, Altitude and Airspeed increasing in deviation from the actual with increased workload is supported. Therefore if pilot workload increases during flight operations, it is likely there will be deviations in airspeed, altitude and heading. This also meaning the above data set rejects the null hypothesis.
Table. 4 - Flight Control Performance
|Table. 4 - Flight Control Performance|
|Workload Level||Communications Signal Quality||Mean Altitude Deviation (Feet)|
When the experimenters analyzed their data for Speech Intelligibility, the null hypothesis of headset configuration used by the pilot groups and the requested repeats from Air Traffic Control (ATC), illustrated there was no significance within the pilot groups. Therefore if considering the type of headset and the observed requested read backs the null is supported; meaning the types of headset had no effect with respect to the number of requested ATC read backs. However when the flight workload was increased and the communications signal quality decreased, the data indicated a statistical significance, inferring that; when pilots workload was increased and the communications quality decreased there was an observed increase in requested ATC read backs. Thus meaning the hypothesis is supported; inferring that pilots are likely to request ATC read backs during increased workloads and poor communications quality. An interesting aspect of the experiment found that, within the group of pilots the Hearing Waived Group requested more ATC read back while wearing the passive foam earplug. This particular observation was not observed within the Non-Hearing Waived group. Therefore the experimenters recommend any hearing waived pilots should use headsets equipped with ANR or CEP passive equipped headsets.
Table. 5 - Speech Intelligibility
|Table. 5 - Speech Intelligibility|
|Workload Level||Communication Signal Quality||Mean Number of Read Backs|
In the eyes of Kristen L. Castr and John G. Casali the case study was a success with positive outcomes. Positive because in the essence of flight safety and flight crew management Human Factors are becoming more apparent and important considerations in the eyes of Airworthiness Authorities. The study found that the choice of hearing protection and communication system are important factors during operational flight. The study identified various workload combinations that affected the Armys helicopter pilots; especially those with hearing waived considerations. However one distinction that is perhaps overlooked is not all aircraft are rotary wing. Therefore do the arguments and the statistics vary positively and/or negatively with respect to fixed wing pilots or pilots within the G.A or Airline industry. The study identified that functional hearing requirements must be satisfied for flight operational duty. One point the experimenters suggested is testing auditory hearing ability while in conjunction with testing motor task activities. This being supported by the deviation increases when workload was increased and communication signal quality decreased.
Want to know more?
- http://hfs.sagepub.com/content/55/3/486 for full Journal.
Contributors to this page
Authors / Editors
PalTechRep - Massey School of Aviation