Psychology, Department of


Date of this Version



Published in Journal of Experimental Psychology: General 143:4 (2014), pp 1429–1436.

DOI: 10.1037/a0036454

PMID: 24684258


Copyright © 2014 American Psychological Association. Used by permission.


A new method combining spatial-cueing and compound-stimulus paradigms draws on involuntary attentional orienting elicited by a spatially uninformative central arrow cue to investigate global/local processing under incidental processing conditions, wherein global/local levels were uninformative (do not aid performance) and task-irrelevant (need not be processed to perform the task). The task was peripheral target detection. Cues were compound arrows, which were either consistent (global/local arrows oriented in same direction) or inconsistent (global/local arrows oriented in opposite directions). Global/local processing was measured by spatial-cueing effects (response time [RT] difference between target locations validly cued by an arrow and targets at different locations), with the test of global/local advantage represented by the effect of cue-level for inconsistent cues (RT difference between global-valid and local-valid cues). Cue-target interval (stimulus-onset-asynchrony [SOA]) was manipulated to test whether global/local advantage varied with relative stimulus availability. Experiment 1 observed a Cue-Level × SOA interaction such that an early, large global cueing effect was followed by a later, smaller local cueing effect, indicative of a global-to-local shift in advantage. This occurred despite knowledge that global/local arrows were uninformative and task-irrelevant and could therefore be ignored, thus displaying key properties of an involuntary process. Experiment 2 added neutral cues (arrow at one level, rectangle at the other) and determined that the reversal was not due to inhibition of the globally cued location or to attenuation of global information but rather to the presence of conflicting spatial information. Experiments 3 and 4 ruled out alternative accounts for these results. These data indicate global precedence in attended but incidentally processed objects.