P03.01 Cognitive processes and Psychological Factors Research

OBJECTIVE

The main objective of this AUTOPACE task is to propose an ATCo Psychological Mental Workload Model that could be used to predict the effects of automation on the ATCo mental workload, Situational Awareness (SA) and performance in the future operating scenarios. Then, the conclusions of this task will be uses as an input for the Training Strategy design together with the risk identified in WP4.


METHODOLOGY

The work of P03.01 Cognitive processes and Psychological Factors Research has been developed through several steps:

Step 1 – Set the principles of an ATCo psychological model in terms of the functional structure of the human cognitive system and the resources necessary for its functioning.

Step 2 – Describe ATCo activity in terms of what is relevant for the psychological assessment to be performed.

Step 3 – Estimate how the cognitive processes (perception, central processing and execution) are expected to be used in the Future Automation Scenarios and compared these cognitive resources with the current ATC paradigm. To develop this step, it has been employed a MWL Computational Model called COMETA.

Step 4 – Explore how the mental workload and the cognitive processes would be modified by the different levels of arousal and ATCo expectations produced by the level of automation.

Step 5 – Propose a preliminary experimental plan to validate the proposed hypothesis.


ATCo PSYCHOLOGICAL MODEL

To describe the structure of the cognitive system we have investigate several psychological models. Finally, the one that we have taken as reference is the proposed by Histon and Hansman (2008). This model stablish that the information that the ATCo receives and that comes from the traffic situation is processed by combining it with the one that the ATCo has stored in its memory after its learning and experience obtained during in the control task execution.

To describe the functioning we employ the called resources, which could be avalible or demanded. On the one hand, available resources are those resources that the ATCo has and, on the other hand, demanded resources are those required to perform the task. In these terms, Mental Work Load can be described as follow:

  • MWL=1 ⇒ The task can be performed optimally
  • MWL>1 ⇒ Mental Overload. The task cannot be performed optimally
  • MWL<1 ⇒ Mental Underload. The task can be performed optimally but could produce boredom and distraction

ATCo TASK MODEL

In addition to the Psychological Model, we have to define the ATCo task in terms that can be assessed with the cognitive model. AUTOPACE uses four main elements to identify and characterise the ATCo task:

  • Typical Control Situation, which allows to identify and characterise a task
  • Control Event, in order to identify the situations that will trigger a task
  • ATCo actions. Description of the minimal unit of observable ATCo behaviour
  • Behavioural primitives, which are used to identify the demand of the cognitive processes required by an action

MAIN RESULTS OBTAINED
  • Effects on available resources

Firstly, the introduction of automation has changed the distribution of the cognitive system, focusing the effort in mainly comprehension and projection. Also, the introduction of automation has shown a more robustness behaviour of mental workload with the same sample of traffic comparing the medium automated scenario with the high automated. Finally, it has been identified that, without taking any mitigation action, non-nominal situations can be unaffordable, especially in case of Conflict resolution failure.

  • Effects on demanded resources

We have stated several hypotheses on how automation impacts the available pool of resources and their allocation to the main task. Some of these hypothesis are, for example, that trust decreases when the level of automation increased or, also,  that in High automation the ATCo will be less engaged with the task allocating less available resources to the task than in medium scenario.

 

If you want to go deeply into the task, have a look to the deliverable here:

D3 1 ATCo Psychological Model With Automation