Multi-tasking & cognitive capacity

• In the topics covered so far in the course, experiments typically involve participants performing one task at a time for many minutes, but in real life, we often perform several tasks at once or alternate rapidly between tasks
• Technology tends to increase the requirement for multitasking
• Experiments have been conducted where participants are required to try to do two tasks at the same time, such as using a mobile phone while driving
• There are 3 types of account of capacity limitations when it comes to multitasking
• Competition for specific resources

  • Tasks might not easily be performed concurrently if they need the same sensory system or effector system
  • Retention of visuo-spatial images is interfered with by an auditory spatial tracking task
  • We cannot comprehend one speech message while shadowing another
• Dual-task performance can be limited by competition for specialised processes or working memory capacities
• In Kleiman’s experiment on access from text to meaning, a concurrent shadowing task designed to "use up" phonological processing resources did NOT interfere with a semantic matching decision for written words
• Debate about dual-task interference has been about whether it results from the need to share the services of a mental general-purpose "central processor" or processing capacity
• Broadbent and Posner postulated the existence of a general-purpose central processor needed for high-level cognitive functions
• According to Broadbent, the selective filter prevents overload of the general-purpose processor
• Other theorists, like Kahneman, postulated a pool of processing resources that could be divided among concurrent high-level processes
• Tasks of any complexity should require the services of a central processor
• Complex tasks should be impossible to combine with any other task of equivalent complexity when both are carried out at near their limiting rate, or performance should show marked tradeoffs
• In the 1970s, certain pairs of complex tasks could be carried out simultaneously without detectable mutual interference or performance tradeoffs
• Complex tasks seem to require different kinds of processing, so they do not compete for domain-specific specialised processing resources
• Concurrent performance of two relatively well-practiced complex tasks seems possible in cases where the tasks use non-overlapping sets of specialised processing resources
• All dual-task interference can be attributed to competition for specialised resources
• An error in one task can briefly delay responses in the other
• An unexpected hazard can interrupt a non-habitual concurrent task such as conversation with a passenger
• Managing multitasking puts a major load on executive processes
• Practice is an important factor in whether two tasks can be combined
• With practice, tasks become automatic and require less executive control functions
• Initial coordination of elements of a skill requires executive control functions, but as practice develops, the load on executive functions diminishes
• There are studies on combining pairs of tasks that initially are difficult to combine
• The experiment of Hirst et al. involved reading text for comprehension while processing concurrent auditory words with a written response
• The "psychological refractory period" (PRP) experiment involves presenting two stimuli requiring different responses very close in time