Performance consequences of interruptions while executing sequential tasks from memory

点击翻译按钮获取中文摘要

Interruptions during the fulfillment of a given task can significantly impair the performance of the interrupted task. In practical settings, this represents a particular problem if a task requires the sequential application of various steps in a predefined order, such as the application of procedures for safety-critical tasks. In this case, interruptions might cause that certain operations are performed too late, are performed in the wrong order, or are omitted entirely. Reports from domains like aviation or medicine show, that performance consequences like this have repeatedly been involved in severe accidents and incidents. Theoretical models dealing with interruptions conceptualize the cognitive demands caused by interruptions as a (prospective) memory task, in which action goals of the primary task have to be kept active in memory while working on the interruption, and have to be reactivated correctly after the interruption. However, the current knowledge about the relevant characteristics of an interruption and the cognitive processes which determine how fast and correct a primary task can be resumed after an interruption is still limited. This is especially true for interrupted sequential tasks (procedures) which have been considered as part of the interruption research only recently.The current project aims at investigating to what extent the performance consequences of interruptions on the fulfillment of sequential tasks depend on certain characteristics of the interruption task. Specifically, three aspects of interruption tasks will be considered which seem to be relevant from a theoretical and practical point of view: the length and the timing of interruptions, the complexity and type of (cognitive) demands of the interruption task, and whether or not an interruption demands an immediate withdrawal of attention from the primary task. The primary task used for this research requires it to respond to a complex visual stimulus by applying eight different work steps in terms of questions addressing certain characteristics of the stimulus. The work steps have to be carried out from memory in a predefined order. Memorizing the questions and their sequence is supported by the keyword WORTKLAU. Thereby, the task simulates typical requirements for the processing of memorized procedures in applied domains. An English version the task was first introduced as research paradigm for interruption research in 2014, and has been adapted and modified for this project. By analyzing processing times and errors rates for different work steps before and after an interruption, performance consequences of interruptions can be characterized in detail and can also be described precisely in their time dynamics. It is expected that the project provides new insights into performance consequences of task interruptions, which are relevant for both, our theoretical understanding, as well as practical guidance for effective interruption management.