Abstract
An important question in Psychological Science concerns the extent to which human information-processing limitations reflect the capacity of general vs. specific cognitive resources. Here we tested whether perceptual (e.g., attentional blink, AB) and decision-making limitations (e.g., psychological refractory period, PRP) occur due to a general, unified bottleneck. To do this we employed a training approach to assess the specificity of practice required to reduce both these impairments. Our logic was that if both limitations respond to the same training regimen then it would suggest they reflect a common limitation. In contrast to previous research that has examined the relationship between the PRP and AB, a key strength of the current approach was that standard AB and PRP paradigms could be employed (in addition to a hybrid AB-PRP task which has previously been used to test this question). Participants completed a PRP task, an AB task and a hybrid AB-PRP task (speeded Task 1 and delayed Task 2) before being allocated to a relevant training group (T1 practice for all paradigms), an irrelevant training group (comparable sensorimotor training) or a control (no training) group. Training groups undertook ~4000 single-task trials across two weeks. At re-test, only the relevant training group showed reductions in PRP and hybrid AB-PRP magnitudes. However, both the relevant and irrelevant training groups showed a significant reduction in AB magnitude relative to controls. This indicates that while task-specific sensorimotor training is required to reduce the magnitudes of the PRP and hybrid AB-PRP, general sensorimotor training attentuates the AB. The findings suggest that while there is overlap in the bottlenecks underlying perceptual and decision-making limitations, the AB also represents a distinct bottleneck in information processing. These findings have implications for understanding the cognitive architecture underlying human information processing and the extent to which training transfers across tasks.
Meeting abstract presented at VSS 2012