Analysis Of Decision-Making Actions Of Crew On Us Flight 1549
The following essay attempts to analyse the decision-making actions taken by the flight crew onboard US Flight 1549 led by expert pilot, Captain Sullenberger. Through the exploration of both Recognition Primed Decision Making Model (RPD) and the Template Theory, it is evident that both model and theory focus and build upon the ideals of past experiences and intuition to resolve complications. It is also agreed by both theory and model that familiarity, options and evaluation are crucial elements for making informed choices. Although the backbone of the models differ slightly, they allow us to decipher the steps taken and attributes required by an expert pilot in order to recover from an incident.
The cognitive process of Decision-Making is to identify problems and consequently formulate appropriate responses when exposed to ill-fated and uncertain situations (Orasanu & Connolly, 1993). However, when errors are made, they may contribute to undesirable outcomes as seen within the aviation industry, where it has accounted for more fatal accidents than perceptual or action errors (Jensen & Benel, 1982). Thus, it has been even theorized by some that the human aspect, the flight crew themselves are to blame as they are the "weak link" within the chain (Feggetter, 1982). As pilots are humans, they are inevitably prone to errors, nevertheless, this can be overcome by training and understanding the procedures taken by experts, in order to decrease or even mitigate the risks. Formally, experts are individuals who are capable of offering strong justifications for a range of queries within a certain domain (Weinstein, 1993). Thus, an expert pilot is characterised by acquiring 1,000 flight hours (Wiggins & O’Hare 2003) while also possessing high levels of motivation, superior learning, performance skills and is adaptive to external dynamics within their environment. Ultimately, the following essay attempts to explain the role played by an expert pilot throughout US Airways Flight 1549 utilising decision-making models such as the Recognition-Primed Decision Model and the Template Theory.
An incident to illustrate the following models can be seen through actions taken by flight crew onboard US Airways Flight 1549. A routine flight from New York's LaGuardia Airport to Charlotte Douglas International Airport ended in disaster after the Airbus A320 suffered dual engine failure due to geese ingestion during the initial climb out phase. According to records obtained from US Airways, Captain Sullenberger had accumulated 19,663 total flight hours before the incident, well above the 1,000-hour mark and thus considered an expert. This is further enforced by the extensive knowledge and experience of the captain as he had 4,765 hours on the Airbus A320 and 8,930 hours as pilot in command (PIC) (NTSB, 2009). By making the difficult decision of ditching into the Hudson River, it ultimately saved everyone onboard. Speaking to CBS reporter Kate Couric, Captain Chelsey Sullenberger stated "For 42 years, I've been making small, regular deposits in this bank of experience, education, and training. And on January 15, the balance was sufficient so that I could make a very large withdrawal” (Couric, 2009, p. 1). Inevitably, through his actions; developed from extensive training and recollection of prior knowledge, was he able to successfully avert disaster.
Naturalistic Decision Making is concerned with the ability to recognize cues prompted from prior experience. Thus, the Recognition-Primed Decision Model (RPD) developed by Klein (1989), builds upon the idea of previous experience and expertise, allowing an individual to plot out a plausible course of action (COA) to successfully execute a plan. It thoroughly relies on one's ability to classify tasks as familiar or unprecedented (Klein & Crandall, 1990) in contrast to deliberately and methodically comparing alternatives using abstract evaluation methods. The classification of tasks is enhanced through the use of heuristics, based on mental shortcuts. Expert pilots consolidate and conceptualise scenarios based on past experiences whereby their knowledge and depth of content has been stabilised overtime (Mogford, 1997). These characteristics are synonymous with expert pilots and correlate to increased situational awareness and informed decision making throughout various phases of the flight (Endsley & Jones, 2004).
In accordance, by familiarizing with the surrounding environment, options are generated serially whereby the most appropiate solution is considered (Klein & Crandall, 1990). During the unfolding situation, options are tested for adequacy, identified for weaknesses and then dealt with. This is fairly evident during US Airways Flight 1549, when Captain Sullenberger realised the options he had, to either land at Teterboro, LaGuardia or the Hudson River. As they were descending at a rate of 1650ft per min (500m per min) from 2818 ft above ground level (AGL) located within downtown New York, the risk was too great, as persistence towards the airport would have resulted in higher fatalities. This seemingly leads onto the practice of evaluation whereby the use of mental stimulation allows one to visualise options implemented within an environment and allow them to make improvements if barriers are detected (Klein & Crandall, 1990). If so, the decision maker is alerted and responds with an appropriate action to challenge the corresponding dynamics, thus the option chosen by Sullenberger to land in the Hudson River rather than either airport. The following model succeeds in its ability to be understood intuitively by experienced individuals however with that being said, experience takes time and thus is impractical for junior individuals. It is also considered ineffective when an individual is placed under stressful situations.
Similiarly, the Template Theory developed by Gobet & Simon (1996) also advances the idea of recognition, based on past experiences. Visual and echoic data obtained from sensory organs such as the eyes and ears are deposited as “templates” and stored within long term memory. Once the indivdual perceives something as familiar, their sensory organs are stimuated, with new information encoded, compared and matched with templates already stored within long term memory (Pi et al. , 2008). As every template has a sentimental value towards the individual, different patterns will be treated differently, in order to successfully differentiate and distinguish one memory from another. Accordingly, as experts possess greater cognitive capacity compared to that of novices they can handle greater volumes of information and thus can quickly choose from an array of data already stored as templates without being overwhelmed (Gobet & Clarkson, 2004). A disadvantage of the theory is that an appriopriate template must be already store within long term memory before an pattern is recognised and action taken. This decreases the holding capacity of the brain, making it less flexible and occupied thus becoming more prone to interference and decay (Gobet, 1997).
Thus, the Template Theory recognises the importance of experience when making accurate decisions and like the RPD model, it also requires familiarity and prior experience from the flight crew. As “templates” are acquired from past experience, Captain Sullenberger in particular would have compared incoming data detected through visual and acoustic stimuli with data already stored within his long-term memory before finding a match. According to Wiggins and O’Hare (2003), information that could then assist in formulating an appropriate decision would be extracted by experts as compared to novices which would tackle the problem as presented. Sullenberger could then have remembered vital memories from his time as a glider and jet pilot in the military before executing his judgement. Mandatory recurrency checks usually issued every 6 months, would have also allowed Sullenberger to rehearse and revise over his training through the recollection of stored templates thus the decisions made were unhurried, non-impulsive and adaptive to him. This is observable when Sullenberger recalls past simulator training and undergoes procedures to restart the Auxiliary Power Unit (APU). Without success he then directs First Officer Skiles to operate the Quick Reference Handbook (QRH) and complete the Engine Dual Failure Checklist. However, as they were below the optimum relight airspeed of three hundred knots they could not restart the engines (NTSB, 2009).
As no simulator could replicate a water-based landing and the only training received was a theoretical class-based lesson, Sullenberger would have drawn inferences and ideas based on intuition and recalled templates regarding past glider experience, allowing him to make the best possible decision. This reminiscent to the evaluation step of the RPD Model where one is alerted when barriers are detected i. e. inability to restart engines, and options reconsidered, i. e. glider instincts kicking in from knowledge retrived from LTM. Ultimately both models agree in accordance that experience is paramount for the safety of a flight.
Ultimately, the Recognition-Primed Decision Model developed by Klein (1989) and the Template Theory developed by Gobet & Simon (1996), provides a deeper insight into the methodology used by an expert pilot when making decisions. The Template Theory highlights the value of past experiences, allowing for a quick and inituitive process of comparing new data with the old. This is similar to the level of knowledge required by an expert within the RPD model, possessing a certain degree of confidence when plotting a COA. As of their similiarities, both models identify the importance of considering options and being alert, allowing for a clear mental space, devoid of other external conflicting factors. However, future prospects show a trend occurring whereby the use of automatic cues will replace vigilant information processing (Mosier et al. , 1998), influenced by advancing technology. The lack of manual flying due to automation bias has ultimately resulted in degraded performance as pilots are trained to actively interpret cues from flight instruments. Expert pilots readily exploit it by matching patterns or combinations which they believe are the most valid and reliable. But because automation has been advertised as being "superior" than that of a human operator there has been a shift towards a reliance on technology, reducing pilot cognitive proficiency which is ultimately a disaster in the making.
