Answers are grounded in this paper and cite their sources.
AbstractSeveral behavioural studies have suggested that rarity is critical for enabling irrelevant, salient objects to capture attention. We tested this hypothesis using the N2pc thought to reflect attentional allocation. A cue display was followed by a target display in which participants identified the letter in a specific colour. Experiment 1 pitted rare, irrelevant abrupt onset cues (appearing on only 20% of trials) against target-relevant colour cues. The relevant colour cue produced large N2pc and cue validity effects even when competing with a rare, salient, simultaneous abrupt onset. Similar results occurred even when abrupt onset frequency was reduced to only 10% of trials (Experiment 2). Experiment 3 examined rare, irrelevant colour singleton cues (20% of trials). Despite being rare and salient, these singleton cues produced no N2pc or cue validity effect indicating little attentional capture. Experiment 4 greatly increased colour cue salience by adding four background boxes, increasing colour contrast and tripling the cue display duration (from 50 to 150 ms). Small cue validity and N2pc effects were obtained, but did not strongly depend on degree of rarity (20% vs. 100%). We argue that rarity by itself is neither necessary nor sufficient to produce attention capture.Keywords: Attention captureN2pcVisual attention We thank Addie Johnson and two anonymous reviewers for comments on earlier versions of this manuscript. We also thank Andrew Morgan for technical support.This research was supported by funding from Oregon State University Undergraduate Research, Innovation, Scholarship and Creativity: START and graduate assistantship from the School of Psychological Science to Birken Noesen.We thank Addie Johnson and two anonymous reviewers for comments on earlier versions of this manuscript. We also thank Andrew Morgan for technical support.This research was supported by funding from Oregon State University Undergraduate Research, Innovation, Scholarship and Creativity: START and graduate assistantship from the School of Psychological Science to Birken Noesen.Notes1 The time window for the N2pc effect typically ranges from 150 to 300 ms after stimulus onset, but the exact timing depends somewhat on stimulus display and task demands (see Luck & Hillyard, Citation1994; Woodman & Luck, Citation2003). Our cue display and target display contained different objects, so one cannot assume identical N2pc time windows. Furthermore, the time course for the target-elicited N2pc also depends on preceding cue conditions. Thus, we used different time windows to assess the cue-elicited N2pc effect and the target-elicited N2pc effect based on when the effect was strongest in the aggregate data.2 The target-elicited N2pc effect analyses revealed a significant effect of cue/target spatial relationship (same hemifield vs. different hemifields) due to how we defined contralateral and ipsilateral ERPs in calculating the N2pc effect (i.e., with respect to the cue location). We repeated the analysis of target-elicited N2pc effect coded with respect to the target location rather than cue location. Results showed that the main effect of cue/target spatial relationship was significant, F(1, 14) = 34.34, p < .0001, . A similar result was observed in Experiment 2, F(1, 20) = 43.64, p < .0001, . The observed pattern – a smaller target-elicited N2pc effect for the same hemifield than different hemifield conditions – was as expected. Given that attention has already been allocated to the relevant colour cue, a target appearing in the same hemifield did not require as much of an attentional shift as a target appearing in the opposite hemifield.3 We reported the mean RT and PEs for each cue condition averaged across groups in Table 3 for ease of readability. More detailed data, broken down by target colour and irrelevant cue colour, can be found at http://people.oregonstate.edu/∼lienm/RarityCaptureSupplement.pdf.4 We repeated the analyses of target-elicited N2pc effects with respect to target location and found that the main effect of cue/target spatial relationship was small and only marginally significant, F(1, 20) = 3.95, p = .061, . Note that because attention was not captured by the irrelevant colour singleton cue, one would not expect attention allocation to the target (i.e., the target-elicited N2pc effect) to vary as a function of where the cue appeared in relation to the target.5 We did not conduct an overall ANOVA including experiment (Experiments 4A vs. 4B) as a variable since the level of cue condition was different between these two experiments (only 4A has a no-cue condition). Nevertheless, a between-experiment comparison on cue validity effects was reported in the Discussion section of Experiment 4.6 We repeated the analyses of target-elicited N2pc effects with respect to target location and found that the main effect of cue/target spatial relationship was not significant in both Experiments 4A and 4B, Fs < 1.56, ps = .2284. The capture by the irrelevant colour singleton cue occurred early during the time window 170–210 ms after cue onset. The increased duration between the cue onset and the target onset (250 ms) in Experiment 4 would allow attention to reallocate back to the central location prior to target onset. Therefore, one would not expect attention allocation to the target (i.e., the target-elicited N2pc effect) to vary as a function of where the irrelevant colour singleton cue appeared in relation to the target location.
No takes yet. Share an insight, caveat, or question.
Noesen et al. (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: