Changes in task demands can trigger a complex and dynamic self-regulatory process which influences a range of metacognitive and physiological states. Individuals continuously monitor task cues, the level of task performance and other internal states in order to form an assessment of current task demands. Cognitive-energetic theories propose that individuals possess a finite amount of available resources which can be allocated to meet task demands. Increased task demands will generate an increase in the level of resources allocated in order to maintain task performance, but only up to a point of maximum effort which is determined by the level of available resources. However, while the level of available resources is finite, it is not necessarily fixed. It is generally held that the sustained allocation of resources depletes the level of available resources, so that prior high task demands will reduce the level of available resources and therefore the maximum level of resources allocated to meet current high task demands. Reduced levels of available resources may also cause a compensatory increase in the level of allocated resources at low demand levels in order to protect against possible performance lapses caused by low resource availability.
The current thesis proposes that, instead of a general process by which all resource allocation depletes the level of available resources, the allocation of attentional control resources depletes the level of available resources but the allocation of information processing resources provides an opposing short-term increase in the level of available resources. This proposal provides a way of accounting for potentially contradictory empirical data and integrating ego depletion theory and malleable resources theory. The thesis develops a resource-based self-regulatory dynamic control model of the human response to task demands in which the level of available and allocated information processing and attentional control resources both influence and are influenced by current and prior task demands. The model also identifies how the level of available and allocated resources contribute to the metacognitive states of perceived difficulty, effort, activation and valence and the physiological state of pupil diameter. Three experiments were conducted to test predictions arising from the model. Experiments 1 and 2 manipulated the level of task demands within a range of relatively simple, short-term, intermittent cognitive and motor control tasks and Experiment 3 manipulated demand level within a sustained, continuous, and complex control task in order to identify the validity of the proposed model under a range of task conditions.
The experiments provided mixed support for the model. The proposal that attentional control demands and information processing demands had opposing effects on the level of available resources was broadly supported by the empirical data which suggests that resource theories need to distinguish the effects of these two types of task demands. However, the level of available resources did not have a simple or consistent effect on the level of allocated resources across the three experiments which highlights the role of strategy in resource allocation and suggests that it may not be feasible to identify a general form for this relationship. Only mixed support was found for the proposed relationships between available resources, allocated resources and the metacognitive states of difficulty, effort and valence which suggests that the levels of available and allocated resources may have only a weak effect on the metacognitive states which appeared to be more strongly influenced by task characteristics.