Printer Friendly

Psychological factors affecting pilots' decisions to navigate in deteriorating weather.

Poor pilot decision-making in deteriorating weather is the leading cause of a significant percentage of fatalities arising from aviation accidents in the last two decades. Research has identified psychological factors underlying pilot decision strategies as the primary reasons for faulty decision making during inclement weather conditions. In the present review, the authors attempt to evaluate existing experimental data on the cognitive and affective processes that govern pilot decision making in changing weather conditions, with specific emphasis on instances of VFR (visual flight rules) flight into IMC (instrument meteorological conditions). We present a consolidated model of pilot decision processes and at each stage in the decision tree we discuss the possible intrinsic and extrinsic factors that might affect the efficacy of cockpit decisions. Based on this model, we examine interventions aimed at reducing the incidence of sub-optimal pilot decision making under poor weather conditions. Suggestions for improving the quality of aeronautical decision making through the use of technology and training are provided.

Ineffective or inappropriate pilot decision-making in poor weather conditions accounts for a significant proportion of fatalities arising from General Aviation (GA) accidents (Wiggins, Martinussen, & Hunter, 1999). According to data from the Aircraft Owners and Pilots Association (1996), despite the introduction of a number of safety-related initiatives over the last twenty years, a significant proportion of GA accidents continue to involve visual flight rules (VFR) into instrument meteorological conditions (IMC; a clearer description of these terms is provided below). An analysis of the National Transportation and Safety Board's (NTSB) aviation accident database indicates that in the past decade, 2.5% of the more than 14,000 GA accidents were classified as involving pilots' decisions to navigate in inclement weather conditions (Goh & Wiegmann, 2001). Whereas the fatality rate of other GA accidents that did not involve weather related accidents was 18%, 75% of all VFR flight into IMC accidents in the same 8-year period was fatal (Goh & Wiegmann, 2001).

The enormity of the fatality associated with weather-related aviation accidents raises the need for research efforts that analyze and develop interventions aimed at mitigating the incidence of such accidents in GA. In response to this need, several researchers over the last decade (e.g., O'Hare, 1992; Driskill, Weismuller, Quebe, Hand & Dittmar, 1995; Goh & Wiegmann, 2001) have empirically examined various aspects of VFR into IMC accidents and have postulated several factors that might assist in reducing the degree of associated fatalities. However, interventions in this area are limited by the lack of a theoretical framework that integrates existing research efforts into a synthesized model of the causes and solutions for weather-related GA accidents.

While some isolated attempts have been made to categorize pilots' decision processes into individual frameworks, no research so far has tried to integrate these efforts into one synthesized model. This is important in that pilot decision making is largely influenced by multiple factors that act on the psychological processes of the pilot simultaneously. Therefore, a depiction of the various stages of decision making along with the extraneous factors affecting these processes is critical to the understanding of the causes and consequences of decisions to fly VFR into IMC. The purpose of this review, therefore, is to develop

a consolidated framework of factors that influence incidences of VFR into IMC, which will serve as a guide to future researchers in the field. Specifically, we evaluate and synthesize existing research attempts and experimental data on the psychological processes which govern pilot decision making in changing weather conditions. We begin with general definitions of VFR into IMC and their associated problems. We then discuss the psychological (cognitive and motivational) factors that affect the choice of a particular flight strategy in inclement weather by relating and integrating three existing models of decision making into a consolidated framework. Finally, we use our decision making model as a guide to generate suggestions for improving the quality of pilot decision making through the use of technology and training.

Definitions of VFR into IMC and associated problems

Visual flight rules (VFR) are a set of aviation regulations under which a pilot may operate an aircraft independently if weather conditions are sufficient to allow the pilot to visually control the aircraft's attitude, navigate, and maintain separation with obstacles such as terrain and other aircraft. Under VFR, the pilot generally controls the attitude of the aircraft by relying on what can be seen out the window, although this may be supplemented by referring to the instrument panel. A pilot flying under VFR is usually required to stay at least a specified distance away from clouds and must stay in areas where the visibility meets minimum requirements. There may be other requirements which vary by country, such as not flying over a solid layer of clouds, or not flying at night.

Under VFR, the pilot is solely responsible for seeing and avoiding other aircraft, terrain, and obstructions such as buildings and towers. Being in contact with air traffic control (ATC) is optional in most airspace under VFR, and the pilot is has the option of selecting the course and altitude to be flown even when in contact with ATC. The pilot may navigate either visually, or by reference to instruments and electronic aids to navigation. The minimum meteorological requirements for VFR are called visual meteorological conditions (VMC). If they are not met then the flight must be flown under instrument flight rules (IFR), i.e., the pilot must have an instrument rating and meet recency of experience requirements pertaining to instrument flight, and the aircraft must be equipped and type-certified for instrument flight.

Unfortunately, statistics indicate several instances of pilots intentionally or otherwise using VFR to fly into IMC conditions, without the appropriate certifications and regulations necessary to navigate in inclement weather. A 1989 NTSB report, based on 361 general aviation accidents occurring between 1983 and 1987 in which VFR flight into IMC was a probable cause or related factor, found that 97% of the possible causes were attributed to the flight crew, and 42% of these cited the manner in which weather information was not obtained (or poorly obtained), and assimilated (Driskill, Weismuller, Quebe, Hand & Dittmar, 1995). An earlier review of aircraft accident data (Jensen, 1982) concluded that 80 to 85% of aircraft accidents can be attributed broadly to pilot error, out of which 51.6% of the fatal and 35.1% of the non-fatal accidents resulted from failures in decision making.

Due to the prominence of pilot error and frequency of VFR flight into IMC as causes of GA accidents, the Federal Aviation Administration (FAA) sponsors several training interventions which address both cognitive and affective components of pilot decision making (Brecke, 1981; Jensen, 1982). The emphasis on the psychological aspects of decision making stems from the fact that explanations for why pilots perform risky actions in response to deteriorating weather conditions (such as VFR flight into IMC) have primarily pointed to issues related to faulty pilot decisions and the general recognition that good decision-making is necessary to maintain safety in aviation (O'Hare, 1992). This recognition, in turn, is engendered by findings in which fatal aviation accidents involve decision errors more than minor accidents that tend to be associated more with procedural execution errors (O'Hare, Batt, Wiggins, & Morrison, 1994; Wiegmann & Shappell, 1997).

In the ensuing sections we discuss in detail three organizing frameworks of decision making that have addressed various psychological aspects that affect decisions to fly in inclement weather conditions. However, the primary shortcoming of these frameworks is that each model individually fails to provide a consolidated picture of all the factors that influence pilot decisions, which is essential to developing a good theoretical as well as conceptual understanding of the causes and consequences of VFR into IMC. In this paper we circumvent this problem by presenting an integrated version of the three models and addressing the manner in which the model relates to the various interventions that have been/need to be incorporated in order to reduce the percentage of fatalities arising due to impaired pilot decision making. This integrated model is an improvement over existing models in that it addresses (a) the basic stages of human information processing, while relating it to (b) the stages of pilot decision making in a situation of inclement weather patterns, and, (c) the psychological (cognitive/affective) factors that affect decision making at each stage. Our contention is that this provides a clearer picture of the psychology governing VFR into IMC decisions in successive stages, which the previous models have failed to elucidate individually.

Organizing frameworks of pilot weather-related decision making

We begin with an understanding of the precise judgment and decision tasks being performed by the pilot, particularly when encountering unfavorable weather conditions. Jensen's (1995) Judgment model (also discussed by Goh & Wiegmann, 2001) describes the multiple stages involved in the judgment process that might take place in aeronautical decision-making situations, such as the decision to continue VFR flight into IMC. Each of these stages is described as below and is illustrated in Figure 1:


a) Problem vigil--pilot maintains a constant state of vigil so that changes in the environment (e.g., appearance of dark clouds) can be detected.

b) Recognition.--pilot realizes the changes in the environment could affect the safety of the flight (e.g., development of dark clouds could lead to a severe thunderstorm)

c) Diagnosis.--pilot attempts to understand the nature of the problem (e.g., assess the possibility that cloud development could lead to a potentially dangerous storm and its implications for flight)

d) Alternative identification--pilot identifies various alternatives (i.e., the choice of an alternative flight path)

e) Risk assessment--the pilot tries to determine the risks associated with the alternatives (gains vs. losses)

f) Background factors--the pilot's decisions are influenced by personal and social pressures

g) Decision-making--the pilot chooses the final course of action (e.g., whether to continue VFR flight into IMC or to divert).

h) Action--the pilot applies the decision by moving flight controls, interacting with passengers, Air Traffic Control (ATC) and other crewmembers.

While the above model provides an elaborate look at the tasks being performed by a pilot, the components of the model can be further simplified and mapped onto Wickens and Hollands' (2000) information processing model of decision making. According to this model (incorporated in Figure 1), the process of decision making can be roughly classified into three sequential categories: information acquisition, situation assessment, and choice of action. Relating this model to the earlier model of pilot decision making (Jensen, 1995), the stage of information acquisition encompasses the first two stages of Jensen's model, namely problem vigil and recognition. In this stage the pilot actively seeks cues from the environment and performance is primarily driven by attention, concentration and perception. Performance at this stage is typically affected by biases such as the salience bias wherein the pilot is mislead by salient but wrong cues, or the confirmation bias wherein probabilistic reasoning leads to adhering to wrong cues.

As depicted in Figure 1, the second stage in the Wickens and Hollands model, situation assessment corresponds to the stages of diagnosis, alternative identification and risk assessment in the Jensen Judgment model, where decision making and the choice of an action is largely influenced by information stored in working memory and long-term memory. Decisions at this stage are most often affected by the overconfidence bias and the availability heuristic (Kahneman & Tversky, 1982), as well as pilots' propensities to cognitively "anchor" (Madhavan & Wiegmann, 2005) onto cues derived in the previous stage. 'Cognitive anchoring' occurs when pilots prematurely make decisions about states of the world (in this case, weather patterns) and rely on these initial decisions to navigate, even if subsequent evidence from the environment contradicts these initial decisions. Such anchoring will negatively affect pilot performance when initial pre-decisions are erroneous. The last stage of the Wickens and Hollands model, choice of action comprises the decision making and action components of the Jensen model (as shown in Figure 1). Decisions at this stage are affected by several "background (intrinsic and extrinsic) factors" as described below.

Background factors contributing to VFR flight into IMC

According to Jensen's (1992) cognitive-affective model of decision-making, successful pilot judgment and decision-making comprises both a motivational and cognitive component (illustrated in Figure 1 with dotted lines). The cognitive component describes the processes by which pilots establish and evaluate alternatives in a decision-making situation. This component invokes the cognitive and discriminative abilities of the individual that depend on human capabilities to sense, store, retrieve, and integrate information (Driskill et al., 1995). The motivational component refers to emotion inducing aspects of decision-making such as gains and losses associated with decision outcomes and social and personal pressures. Based on the cognitive-affective framework, Goh and Wiegmann (2001) have enumerated specific factors that contribute to VFR flight into IMC. These include sub-optimal situation assessment, degraded risk perception, motivation, decision framing and social pressures, which, in turn, can be further classified into cognitive ("thinking-oriented") vs. affective ("feeling-oriented") factors.

A) Cognitive factors

a) Sub-optimal situation assessment. One reason why pilots engage in VFR flight into IMC is a failure to assess hazards (deteriorating weather conditions). In an analysis of flight and flight-related mishaps in the U.S Navy and Marine Corps, Wiegmann and Shappell (1997) found that approximately 22% of accidents that were due to human error resulted primarily from diagnostic errors and these diagnostic-error related accidents were more serious than those related to aircraft-handling errors. Contingent with pilots' low accuracy in assessing situational hazards, there has been growing interest in the cognitive strategies associated in pilot decision-making. One such study by Driskill et al., (1995) required pilots to rank a series of 27 weather scenarios and indicate levels of comfort for each. Most pilots used a compensatory decision-making strategy characterized by consideration of each option in its entirety and the assessment of positive and negative attributes, finally selecting an option with the greatest number of positive features. On the other hand, a non-compensatory approach is characterized by an assessment of each option against a predetermined criterion. Any option that reaches the criterion on any one of the attributes under consideration may be selected (Wiggins et al., 1999).

In order to investigate whether the above situation assessment strategy was uniformly adopted by pilots across different geographical locations, Wiggins et al., (1999) conducted parallel assessments of pilot decision-making strategies in three countries: The United States, Norway and Australia. Results suggested that irrespective of geographical location, pilots tended to perceive a similar level of relative risk associated with various weather-related phenomena such as development of snow, dust and low visibility, regardless of the type of terrain over which the scenario was set. Secondly, inexperienced pilots had a tendency to overestimate their capability to assess conditions in comparison to more experienced pilots. These cultural similarities in situation assessment strategies adopted by pilots have important implications for the generalizability of pilot training systems across a range of environments and terrain types.

b) Degraded risk perception. Decision-making under uncertainty also involves the perception of risk. In the case of VFR flight into IMC, pilots may assess the situation accurately (i.e., detect the deteriorating weather) but may not realize the risks involved in continuing with the flight (Goh & Wiegmann, 2001). Risks are the probabilities of suffering a loss while hazards are the sources of danger and may be classified as to severity of possible outcome (O'Hare, 1990). It is possible to be aware of hazards (accurate situation assessment) without being aware of the risk associated with that hazard (inaccurate risk perception). O'Hare (1990) developed and administered the Aeronautical Risk Judgment Questionnaire (ARJQ) to a sample of licensed pilots to obtain data on pilots' perceptions of their abilities, willingness to take risks, hazard awareness, and judgments of the risks involved. A subset of these pilots was tested on a computerized test of flight decision-making involving a proposed VFR flight into marginal weather conditions. Results from the ARJQ indicated that on average pilots exhibited low levels of risk awareness combined with an unreasonably optimistic self-appraisal of abilities and high scores on a measure of "personal invulnerability" to factors associated with aviation accidents (O'Hare, 1990). This supported earlier findings by the U.K Civil Aviation Authority (1988) who cited the psychological factors contributing to pilot errors in bad weather included "excessive optimism", a "reluctance to admit limited capability, and "lack of appreciation of real dangers" (Goh & Wiegmann, 2001).

Goh and Wiegmann (2001) examined the above factors in a laboratory experiment with pilots, while simulating conditions of VFR flight into IMC in real time. Findings suggested that pilots who continued VFR flight into IMC made errors early in the decision-making process in the form of inaccurate assessments of visibility, and this was compounded by their poor risk perception, overconfidence in their flight skills, and a reduced sense of vulnerability to weather hazards and pilot error.

B) Affective factors

a) Motivation. Another possible reason for why pilots engage in VFR flight into IMC has to do with motivation. Pilots may diagnose and perceive the risks accurately, but other motivational factors impact their decisions (Goh & Wiegmann, 2001). The motivational factors may be the urgency to reach a destination in a particular time period (known as "get-home-itis") or other personal and social pressures. O'Hare and Smitheram (1995) asked pilots to indicate how important eight personal and social factors were in contributing to their decision to continue or divert flight. Their results indicated that social and peer influences have an important role to play in a pilot's decision to engage in risky flights. However, data from existing research suggests the need for more realistic simulated situations to further clarify the relative impact of social and personal pressures on pilot decisions.

b) Decision framing. According to Prospect Theory (Kahneman & Tversky, 1982), a person's choice between a risky and a safe course of action depends on how an option is framed. In the case of VFR flight into IMC if pilots frame the decision to divert a loss (e.g., wasted time, money, and effort), the pilot might tend to be risk seeking and choose to continue with the flight. On the other hand, if the pilot frames the diverting as a gain (e.g., it is safer), he or she would be risk-averse and choose to divert from the flight (Goh & Wiegmann, 2001). O'Hare and Smitheram (1995) conducted a study to investigate the effects of framing on pilots' decision to continue or divert from a VFR flight into IMC situation. They found pilots were likely to continue flight into adverse weather when they were encouraged, via experimental manipulations of the alternatives, to frame continuing VFR into IMC in terms of losses and diverting in terms of gains. This effect did not occur when pilots used their own natural frames.

c) Social pressures. Another extraneous factor that has the potential to affect pilot decision making is the existence of social (or peer) pressure. According to Goh and Wiegmann (2001), in the case of VFR into IMC flights, pilots may be pressured to reach their destination sooner if they have irritable passengers on board. In addition, they may also feel the need to impress passengers with their flight skills. However, the degree to which social pressures might affect pilot decision making still requires further exploration.

As can be seen in Figure 1's depiction of the background factors affecting aeronautical decision-making, factors influencing the early stages of decision-making are primarily cognitive in nature. Specifically, factors related to maintaining vigilance, perception and assessment of risks and hazards draw on the pilot's intellectual and discriminative abilities. On the other hand, our assessment of the factors listed above suggests that the later stages of pilot decision-making are predominantly affective or motivational in nature including the influence of social and personal background factors and decision framing.

Having delineated the psychological factors that potentially affect decision making in deteriorating weather conditions, the next step is to examine the interventions to be taken in order to minimize/eliminate instances of VFR into IMC. An evaluation of the existing technological and psychological training methods for pilot decision making follows, with specific emphasis on the cognitive and motivational vulnerabilities of pilot decision making in inclement weather as discussed above.

Steps toward improving pilot decision making

The framework described above suggests the necessity for effective interventions to improve the quality of pilot weather-related decision making, particularly based on their relevance to improving both pilot cognition and motivation. Such interventions have been attempted over the last decade and can be channeled into three specific categories of tools- Training, Displays, and Automation.

a) Training--Enhancing flight experience

Although experts can also be prone to the potential pitfalls of heuristics and biases, novice pilots can benefit from training in decision making skills. Training interventions typically focus on the early stages of decision making, namely stages that enhance the pilot's ability to acquire and recognize cues in the environment (see Figure 1). In addition, training efforts have been focused on attitude modification--specifically, a study by Lester, Diehl, and Buch (1985) dealt with the teaching and avoidance of attributes such as anti-authority, external control, impulsivity, invulnerability, and machismo. However, it must be pointed out that having a certain attitude does not always lead to possessing its associated cognitive skills especially "at the novice level." Aids such as the DECIDE model (Clarke, 1986) or the ARTFUL decision tree (O'Hare, 1992) involve the use of checklists. However, the problem with checklists is that they can be time-consuming and may not effectively account for individual differences--checklists are not particularly adaptive. Therefore, modern self-paced training systems should allow for the integration of previous research in training, as well as provide an insight into cognitive factors that lead to sub-optimal situation assessment and degraded risk perception so as to pave the way for improvements in the later stages of decision making.

Computer-based tutoring systems--In order to effectively build and enhance the early stages of decision making (as evidenced in Figure 1), attempts have been made to computerize training efforts in weather-related decision-making. The Pilot Judgment Tutoring System (PJTS) teaches key cognitive skills by allowing the user to (1) "recognize the requirement" (corresponding to the stage of 'information acquisition') for (its) application, and, (2) implement the skill in a timely and accurate manner (corresponding to the stage of 'situation assessment'; Wiggins, O'Hare, Guilkey, & Jensen, 1997). To determine the pertinent cues and skills, a collection of expert skills is typically arrived at through Cognitive Task Analysis (CTA) as well as Subject Matter Experts (SME). Each SME is prompted to recall an experience where they had made a "poor" weather-related decision. In addition, an identification of 'expert' strategies and skills was is determined by the SME. Likewise, behavior typical of 'novices' (such as the misapplication of heuristics and biases) is determined. Adaptive approaches need to be integrated into this framework to account for various skill levels and rates of skill acquisition in different pilots (Payne, Bettman, & Johnson, 1993). This allows the user to progress from a more analytical or time-constrained decision policy to one that is more naturalistic, or recognition-based, leading to appropriate assessment of the situation.

Theory to system development--As discussed above, the PJTS allows the fostering of recognition-based decision behavior through the detection of cues related to deteriorating weather conditions, thereby assisting the first two stages of the decision making model. Users first view a sample VFR into IMC accident and are prompted to determine the three leading causes of the accident. One of five meteorological phenomena is selected: stable deteriorating cloud base, embedded thunderstorm activity, isolated thunderstorm activity, severe haze, or low level cloud development (Wiggins et. al., 1997). For each phenomenon, the user may select a 'problem recognition' (Stage 1) phase. In this case, the particular phenomenon is introduced via video and text. Afterwards, various cues are detected and taught through the use of typing in keywords and receiving cards that describe the nature of the cue. A cue-integration stage (Stage 2) follows through the showing of a flight where conditions slowly deteriorate to the point where continued VFR would be deemed unacceptable. The user has to combine the newly-learned cues to determine the point where conditions have become unacceptable.

Along the lines of the model illustrated in Figure 1, the second stage for each meteorological phenomenon relates to a problem-solving stage that is independent of the problem-recognition stage. A practice scenario typically involves a simulated flight where conditions have deteriorated to the point where the user has to consider an alternative destination. The program then asks the user to input the strategy used in making the decision. Feedback is given for the selected decision strategy. Finally, the user experiences a dynamic simulation where conditions slowly deteriorate--all information that would be available in-flight is provided. Periodically, the user is asked to decide (with a time limit) whether to continue the flight or abort. This allows for the integration of both stages (problem-recognition and problem-solving) and the smooth flow of cognitive activity from the information acquisition to the situation assessment stage.

b) Displays--beyond bottom-up processing

While training is an important solution to the development of improved decision making skill sin the early stages, looking for deteriorating conditions is not a pilot's sole task. Workload demands often get to the point where displays of weather conditions may be as helpful (or even more helpful) in detecting and integrating cues than cognitive skill training. Weather announcements are mainly transmitted through the National Weather Service via voice warnings--attention, however, may not be focused on such announcements due to concurrent tasks such as interaction with ATC or navigating the aircraft. Benefits of implementing data link systems include the reduction of demands on working memory as well as the presence of accurate real-time weather data in situations where reliable data collection may be difficult.

Graphical weather service--Research has been done on utilizing data link to visually present pertinent weather information (Dershowitz, Lind, Chandra, & Bussolari, 1995). This digital presentation of information reduces the amount of time and effort spent in cue integration and problem diagnosis by providing pilots with a consolidated 'view of the world' or situation assessment. For example, the Graphical Weather Service (GWS) tracks precipitation for a fixed area of 2 km by 2 km. Severity of precipitation is displayed through the use of colors (green, yellow, and red)--similar to weather displays on the evening newscast. In addition to precipitation, systems called "Advanced Weather Graphics" (Lindholm, 1995) allow for the display of other potentially key weather phenomena. For example, certain products show the presence of storms and lightning (in 3-D) at any selected flight level and time. More importantly, information on jet streams, winds, temperature, and even icing potential can be provided in a dynamic fashion that displays current and predicted conditions. The 2 km by 2 km area provided by GWS can be manipulated to a zoom level appropriate for the user. The primary advantage of this visual representation of events is that it allows the pilot to move to the later stages of decision making involving the choice of action (see Figure 1) quickly without investing to much time and effort on early stage decision making. This is particularly critical in time-pressured situations or in rapidly deteriorating environmental conditions.

Limitations with data link--Despite the above advantages, a variety of considerations must be taken into account when implementing data link. First of all, the current amount of data vs. the bandwidth (resources) available for data often requires the use of compression. Presenting data in the compressed form has been known to create a sense of distortion, leading to increased risk-taking (thereby having damaging effects on later stage decisions) such as entering into a high precipitation area (Dershowitz et. al., 1995). In addition, the head-down time required to visually scan the data pulls attention away from other scanning tasks--a potentially dangerous situation (Dismukes, Young, & Battelle, 1998). In addition, the presence of information (especially in the "Advanced Weather Graphics" systems) at any given time must be relevant to the decision-making process; a careful balance must be done to ensure that neither omission nor clutter of information occurs. In order to compensate for the drawbacks inherent in training interventions and display design, researchers have recently proposed the introduction of "intelligent" automation or decision aiding technology to enhance the quality of cockpit decision making.

c) Automation--an intelligent tutor

On the one hand, display technologies can be argued as a lower-level of automation, wherein the aid collects and displays data in cases where it may be difficult for the user to do so. This significantly enhances the early stages of decision making and paves the way for a more rapid choice of action in later stages. However, higher levels of automation may be beneficial in situations where it is extremely difficult or even impossible for the pilot to independently assess the overall state of the system (in this case, the aircraft in critically deteriorating weather conditions). Such higher-level automated aids such as the GPWS (Ground Proximity Warning System) and TCAS (Traffic Alert and Collision Avoidance System) provide alarms and alerts in the presence of a hazard in the environment, and in some cases even provide recommendations for future action. These automated systems assist, support or even perform the last stages of decision making for the pilot, namely functions associated with risk assessment, decision making and action selection.

Research in this context has been conducted to assess and monitor turbulence, a common meteorological phenomenon. The Turbulence Assessment and Monitoring System (TAMS) is a pilot-centered system which provides guidance and links to key pilot decisions from turning on the seat belt sign to avoiding an area completely. The system uses current sensors and avionics processors, to factor into account the aircraft's attributes. For example, the same perceived turbulence will require different actions--a smaller corporate jet might have to descend to a "calmer" altitude while the crew of a Boeing 747 might have to simply postpone food service. Surveys were performed by Bass and Ernst-Fortin (1991) to determine the criteria for TAMS. Results indicated that current, peak and forecasted turbulence intensity would be helpful as well as information on suggested pilot procedures and frequency of assessment.

Ultimately, automated decision aids can be implemented to improve pilot decision making at various levels, similar to the effect observed in domains such as process control and healthcare. However, the primary problem associated with the implementation of automation in the cockpit is that the 'framing' or presentation of diagnostic information from the aid has a significant effect on information utilization. Improperly presented alerts or unclear information can have counterproductive effects on information utilization and navigating performance (Lacson, Wiegmann & Madhavan, 2005). Furthermore, there are several instances of pilots creatively disabling systems due to problems associated with the false alarm rates of these automated systems (Thomas, Wickens & Rantanen, 2003) which ultimately lead to poor decisions and the choice of wrong actions. Such errors that occur in the later stages of decision making, when undetected in a timely manner, can be even more degrading and dangerous than errors of cognition generated in the early stages of decision making. Therefore, though significant effort has been directed toward supporting and improving the quality of pilot decision making toward reducing instances of VFR into IMC, incidences of VFR into IMC still continue to be a major force in GA accidents today. Clearly, additional research should be conducted to examine the effectiveness of existing measures and the implementation of new techniques to improve the safety of GA, keeping in mind their relevance and relationship to the tasks being performed by the pilot under deteriorating weather conditions (as evidenced by the model presented in Figure 1).


It is evident that while VFR flight into IMC consists of about 4% of all GA crashes, they still constitute 19% of all GA fatalities (NTSB, 1989). As discussed in this article, causes of VFR into IMC conditions include a variety of cognitive factors such as situation assessment and risk perception as well as affective factors that include motivation and decision framing, combined with several biases and heuristics. Decision-making models can help determine the causes of VFR into IMC incidents by delineating the variables that typically affect the quality of decisions at various stages of the navigating process. Consequently, solutions to improve pilot decision making can be found through cognitive aspects of training, displays that allow for easy detection and integration of cues, as well as automated tools to assess and formulate courses of action.

The primary limitation with the integrated framework of cognitive/affective processes described in this paper is the relatively mechanistic nature of the various decision making stages discussed. While the decision models do provide an integrative picture of pilot decision making, it is important to note that pilot behavior, and human behavior in general, tends to be largely contextual in nature and is, on occasion, influenced by intrinsic and extrinsic factors beyond the scope of this model. Such factors include personality and individual differences, moral and religious factors, and unprecedented environmental variables. Therefore, while current research efforts have succeeded in helping develop a better understanding of VFR into IMC, and interventions have led to a reduction in the incidence of weather-related fatalities to an extent, further research should be conducted to design foolproof initiatives which would pave the way for safer skies.


Bass, E. J. & Ernst-Fortin, S.T. (1991). Pilot decision aid requirements for a real-time turbulence assessment system. Proceedings of the International Symposium on Aviation Psychology. Columbus: Ohio State University.

Buch, G. & Diehl, A. (1984). An investigation of the effectiveness of pilot judgment training. Human Factors. 26(5), 557-564.

Clarke, R. (1986). A new approach to training pilots in aeronautical decision making. Frederick, MD: AOPA Air Safety Foundation.

Dershowitz, A., Lind, A. T., Chandra, D.C., & Bussolari, S, R. (1995). The effect of compression induced distortion of graphical weather on pilot decision making. Proceedings of the 8th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Dismukes, K., Young, G., & Battelle, R.S. (1998). Effective management requires a careful balancing act. ASRS Directline: Cockpit Interruptions and Distractions, 10, 2-10.

Driskill, W.E., Weismuller, J., Quebe, J., Hand., & Dittmar, M. (1995). Analysis of the weights applied to weather information by pilots. Proceedings of the 8th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Goh, J., & Wiegmann, D. A. (2001). Visual flight rules (VFR) flight into instrument meteorological conditions (IMC): A review of the accident data. Proceedings of the 11th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Goh, J., & Wiegmann, D.A. (2001). Visual flight rules flight into instrument meteorological conditions: An empirical investigation of the possible causes. The International Journal of Aviation Psychology, 11(4), 359-379.

Hunter, D.R., Driskill, W.E., Weissmuller, J., Quebe, J., Hand, D., & Dittmar, M. (1995). Analysis of the weights applied to weather information by pilots. Proceedings of the 8th International Symposium on Aviation Psychology. Columbus: Ohio.

Jensen, R.S. (1982). Pilot judgment: Training and evaluation. Human Factors, 24, 61-73.

Jensen, R.S. (1995). Pilot judgment and crew resource management. Brookfield, VT: Ashgate., Kahneman, D., & Tversky, A. (1982). Choices, values, and frames. American Psychologist, 39, 341-350.

Lacson, F., Wiegmann, D.A., & Madhavan, P. (2005). Effects of attribute and goal framing on automation compliance and reliance. Proceedings of the 49h Annual Meeting of the Human Factors and Ergonomics Society. Santa Monica:, CA.

Lester, L.F., Diehl, A.E., and Buch, G. (1985). Private pilot judgment training in flight school settings: A demonstration project. Proceedings of the 3rd International Symposium on Aviation Psychology. Columbus: Ohio State University.

Lindholm, T.A. (1995). Advanced aviation weather graphics--information content, display concepts, functionality, and user needs. Proceedings of the 8th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Madhavan, P., & Wiegmann, D. A. (2005). Cognitive anchoring on self-generated decisions reduces operator reliance on automated diagnostic aids. Human Factors, 47 (2), 332-341.

National Transportation Safety Board (1989). General aviation accidents involving visual flight rules flight into instrument meterological conditions. NTSB/SR-89/01, National Transportation Safety Board, Bureau of Safety Programs: Washington, D.C.

O'Hare, D. (1990). Pilots' perception of risks and hazards in general aviation. Aviation, Space & Environmental Medicine, 61(7), 599-603.

O'Hare, D. (1992). The "artful" decision maker: A framework model for aeronautical decision making. The International Journal of Aviation Psychology, 2, 175-191.

O'Hare, D., Batt, R., Wiggins, M., & Morrison, D. (1994). Cognitive failure analysis for aircraft accident investigation. Ergonomics, 37, 1855-1869.

O'Hare, D., & Smitheram, T. (1995). "Pressing on" into deteriorating conditions: An application of behavioral decision theory to pilot decision making. The International Journal of Aviation Psychology, 5(4), 351-370.

Thomas, L. C., Wickens, C. D., & Rantanen, E. M. (2003). Imperfect automation in aviation traffic alerts: A review of conflict detection algorithms and their implications for human factors research. In Proceedings of the 47th Annual Meeting of the Human Factors and Ergonomics Society, Santa Monica: CA.

Vigeant-Langlois, L.N. (1999). Pilot information needs and strategies for operations in icing conditions. Proceedings of the 10th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Wickens, C.D., & Hollands, J.G. (2000). Engineering psychology and human performance (3rd ed.). Upper Saddle River, NJ: Prentice Hall.

Wiggins, M., Connan, N., & Morris, C. (1995). Self perceptions of weather-related decision-making ability amongst pilots. Proceedings of the 8th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Wiggins, M., Martinussen, M., & Hunter, D.R. (1999). Three continents but a single mind when it comes to weather. Proceedings of the 10th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Wiggins, M., O'Hare, D., Guilkey, J., & Jensen, R. (1997). The design and development of a computer-based pilot judgment tutoring system for weather-related decisions. Proceedings of the 9th International Symposium on Aviation Psychology. Columbus: Ohio State University.

Author info: Correspondence should be sent to: Dr. Poornima Madhavan, Carnegie Mellon University, Porter Hall 208-J, Pittsburgh, PA 15213; or

North American Journal of Psychology, 2006, Vol. 8, No. 1, 47-62.

Poornima Madhavan (1) and Frank C. Lacson (2)

Carnegie Mellon University (1)

University of Illinois at Urbana-Champaign (2)
COPYRIGHT 2006 North American Journal of Psychology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Madhavan, Poornima; Lacson, Frank C.
Publication:North American Journal of Psychology
Geographic Code:1USA
Date:Mar 1, 2006
Previous Article:Interviewers' ratings of data quality in household telephone surveys.
Next Article:The marital disillusionment scale: development and psychometric properties.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters