Standard text reading involves frequent eye movements that go against normal reading order. The function of these "regressions" is still largely unknown. The most obvious explanation is that regressions allow for the rereading of previously fixated words. Alternatively, physically returning the eyes to a word's location could cue the reader's memory for that word, effectively aiding the comprehension process via location priming (the "deictic pointer hypothesis"). In Experiment 1, regression frequency was reduced when readers knew that information was no longer available for rereading. In Experiment 2, readers listened to auditorily presented text while moving their eyes across visual placeholders on the screen. Here, rereading was impossible, but deictic pointers remained available, yet the readers did not make targeted regressions in this experiment. In Experiment 3, target words in normal sentences were changed after reading. Where the eyes later regressed to these words, participants generally remained unaware of the change, and their answers to comprehension questions indicated that the new meaning of the changed word was what determined their sentence representations. These results suggest that readers use regressions to reread words and not to cue their memory for previously read words.

The importance of social context in affecting attention has recently been highlighted by the finding that the presence of a passive, nonevaluative confederate can improve selective attention. The underlying mechanism, however, remains unclear. In this paper, we argue that social facilitation can be caused by distractor inhibition. Two distinct sources of evidence are provided from an experiment employing the Stroop task with and without social presence. First, analysis of the response time (RT) distribution indicates that interference is reduced at relatively long RTs. This is consistent with an inhibitory mechanism, whose effects build up slowly. Further support is provided by showing that social facilitation is prevented by using short response-to-stimulus intervals that are thought to reduce cognitive control processes.

No abstract

No abstract

High trait anxious individuals tend to show biased processing of threat. Correlational evidence suggests that executive control could be used to regulate such threat-processing. On this basis, we hypothesised that trait anxiety-related cognitive biases regarding threat should be exaggerated when executive control is experimentally impaired by loading working memory. In Study 1, 68 undergraduates read ambiguous vignettes under high and low working memory load; later, their interpretations of these vignettes were assessed via a recognition test. Trait anxiety predicted biased interpretation of social threat vignettes under high working memory load, but not under low working memory load. In Study 2, 53 undergraduates completed a dot probe task with fear-conditioned Japanese characters serving as threat stimuli. Trait anxiety predicted attentional bias to the threat stimuli but, again, this only occurred under high working memory load. Interestingly however, actual eye movements toward the threat stimuli were only associated with state anxiety and this was not moderated by working memory load, suggesting that executive control regulates biased threat-processing downstream of initial input processes such as orienting. These results suggest that cognitive loads can exacerbate trait anxiety-related cognitive biases, and therefore represent a useful tool for assessing cognitive biases in future research. More importantly, since biased threat-processing has been implicated in the aetiology and maintenance of anxiety, poor executive control may be a risk factor for anxiety disorders.