In previous studies, pilots made roll reversal errors (RREs) when responding to a ‘moving- horizon’ type attitude indicator (AI). It was argued that it was the ambiguity of this display leading to RREs. Later, using non-pilots, it was found that RREs were in many cases caused by expectation-induced misperception of the AI bank angle, which can arise from a pilot experiencing spatial disorientation. The current study evaluated the role of expectation in the control strategy of professional pilots using a hexapod simulator. First, an effective man-in-the-loop spatial disorientation scenario was developed, based on the ‘leans’ illusion. Here, the expectation of ten experienced and eight novice pilots was manipulated with the simulator motion without the AI, after which they had to turn the aircraft back level when the AI was shown again. An error was made in 16.7% of the runs where the bank angle on the AI was opposite to the earlier given physical motion cues, which is 2.9 times more compared to a baseline condition in which no motion, and thus no manipulated expectation, was present. In a third condition where a motion-induced expectation of the bank angle was present, but a level AI was shown later, no errors were made. It was also found that experienced pilots made slightly more errors than novice pilots and that their reaction time was on average half a second higher. It was concluded that an initially induced expectation about the bank angle does not directly influence the control strategy of pilots, but that it does make them more vulnerable to misinterpretations of the AI, leading to RREs being made. This should be taken into account when developing new displays and technologies, as it should at all times guide pilots in their decision-making, minimizing the chance of misinterpretations.

In previous studies it was shown that both pilots and non-pilots sometimes make roll reversal errors (RREs) when the aircraft bank angle is presented on a moving horizon type attitude indicator. This incorrect input has been shown to be caused by a misinterpretation of the horizon symbol on such an attitude indicator. The figure-ground relation of the aircraft symbol and horizon symbol on the attitude indicator has been cited as being a contributing factor to this misinterpretation. A Multi-Layer Display (MLD) is tested as a possible intervention tool for these interpretation errors. With this display type, the horizon symbol is presented on a physically deeper layer, making it possibly more easily interpreted as being in the background, and the aircraft symbol is presented on a second layer on the foreground. A group of 23 non-pilots and 18 pilots were tasked to perform a recovery task on a desktop-based simulator without outside visuals and only an Attitude indicator on both a MLD as well as a conventional Single Layer Display for repeated-measures. Although roll reversal errors were observed at a comparable level to previous research, no significant difference were observed on the error rate between the two display types. A possible explanation is that the stereoscopic depth of the MLD dissipates when the subject sits still and remains exactly in front of the display. This leaves open the possibility that other types of depth cues, such as monocular cues, can be added and used to enhance the figure-ground relation in displays.

Recent studies and accident investigations show the detrimental effects of startle and (automation) surprise on flight crew performance in terms of cognitive reasoning and sensemaking. Previous research conducted at Delft University of Technology shows a positive effect of training variability on crew performance in surprising tasks related to training, but in case of novel, unrelated surprising tasks, training methods remain to be investigated. Other recent studies and proposals suggest to use checklist-based training methods to prepare pilots for puzzling and surprising scenarios, but effects on flight crew performance have yet to be published. Therefore a simulator-based, between-subjects experiment was conducted to explore the effects of a four-item checklist-based training on pilot performance. Training elements focused on managing stress and enhancing situation awareness, as well as improving sensemaking and decision-making processes. It was expected that trained pilots would exhibit better overall performance in test scenarios involving off-nominal, surprising situations. Nevertheless, results indicated no significant difference in pilot response to an initial disturbance, but did show significant performance improvement in scenarios involving a second, ensuing upset compared to the baseline group. This suggests that checklist-based training methods offer a structured problem-solving approach, resulting in better response to consecutive disturbances in select scenarios.

After several recent flight safety events, such as the accident of Air France flight 447 in 2009, investigators determined that surprise and startle can severely disrupt pilot responses. They concluded that pilots need to be better prepared for unexpected and potentially startling situations. In response, aviation safety authorities have recommended and mandated that startle and surprise should receive more attention in pilot training. However, there is insufficient scientific data available on pilots’ behavior in startling and surprising situations, and on how they can best be trained for these situations. This thesis addresses this problem, by studying startle and surprise in pilots, and by investigating which training interventions can strengthen the pilots’ response to unexpected situations...@en