Search

Previous research offers conflicting suggestions about whether "high-priority" verbal stimuli such as an individual's own name or emotionally charged words automatically grab attention and/or can be detected without the usual capacity limitations. Nine experiments investigated this issue, using visual search through displays of words. In speeded search tasks, the subject's own name was detected more quickly than other targets, but in no case were search slopes flat enough to suggest parallel search or "pop-out". Further, names were not found to be unusually potent distractors. Emotionally charged words were neither more readily detected as targets nor more potent as distractors than neutral words. A comparison of observers' accuracy in searching briefly exposed simultaneous vs. successive displays provided further evidence that search for "highpriority" word targets is subject to the same severe capacity limitations as those that are found with search for neutral words. People sometimes seem, rather uncannily, to notice when their name is mentioned in a conversation, even if they were not consciously attending to this conversation. Often, they say, it is as if the name seems to "jump out". Similar effects have been reported with emotion-laden words and voices. The literature on the cognitive processing of such highpriority affective stimuli is somewhat confusing, however, with various conflicting results scattered around the literature. This paper describes a series of experiments undertaken to try to clarify the ways in which high-priority stimuli may be processed differently from other stimuli within the context of a particular task: speeded or unspeeded visual search through displays of words. Aside from their intrinsic interest, the effects of high-priority affective stimuli may shed light on a number of issues. One is the long-running controversy over the extent to which unattended stimuli are processed to a semantic level, as suggested by "late-selection" theories. While many writers have advocated compromise formulations (Johnston & Dark, 1982; Lavie

Bargh et al. (2001) reported two experiments in which people were exposed to words related to achievement (e.g., strive, attain) or to neutral words, and then performed a demanding cognitive task. Performance on the task was enhanced after exposure to the achievement related words. Bargh and colleagues concluded that better performance was due to the achievement words having activated a "high-performance goal". Because the paper has been cited well over 1100 times, an attempt to replicate its findings would seem warranted. Two direct replication attempts were performed. Results from the first experiment (n = 98) found no effect of priming, and the means were in the opposite direction from those reported by Bargh and colleagues. The second experiment followed up on the observation by Bargh et al. (2001) that high-performance-goal priming was enhanced by a 5-minute delay between priming and test. Adding such a delay, we still found no evidence for high-performance-goal priming (n = 66). These failures to replicate, along with other recent results, suggest that the literature on goal priming requires some skeptical scrutiny.

Williams and Bargh (2008) reported an experiment in which participants were simply asked to plot a single pair of points on a piece of graph paper, with the coordinates provided by the experimenter specifying a pair of points that lay at one of three different distances (close, intermediate, or far, relative to the range available on the graph paper). The participants who had graphed a more distant pair reported themselves as being significantly less close to members of their own family than did those who had plotted a more closely-situated pair. In another experiment, people's estimates of the caloric content of different foods were reportedly altered by the same type of spatial distance priming. Direct replications of both results were attempted, with precautions to ensure that the experimenter did not know what condition the participant was assigned to. The results showed no hint of the priming effects reported by Williams and Bargh (2008).