Methods to Test Visual Attention Online

Online data collection methods have particular appeal to behavioral scientists because they offer the promise of much larger and much more representative ...

University of Wisconsin-Madison

When identifying two targets presented in a rapid serial visual presentation stream, one's accuracy on the second target is reduced if it is presented shortly (within 500 msec) after the first target-an attentional blink (AB). Individuals differ greatly in the size of their AB. One way to learn about the AB is to understand what underlies these individual differences. Recent studies have suggested that when a broadened or diffused attentional state is induced, the AB deficit can be attenuated. The present study examined whether natural (dispositional) individual differences in focus and diffusion of attention as assessed by the global/local task could predict performance on the AB task. Performance that was consistent with diffusion correlated negatively with AB size, and performance that was consistent with focusing correlated positively with AB size, showing that dispositional focus and diffusion of attention can predict individual differences in the AB. These findings are consistent with the Olivers and Nieuwenhuis (2006) overinvestment hypothesis.

Individual differences in dispositional focus of attention predict attentional blink magnitude.

When the second of two targets is presented temporally close (within 500 ms) to the first target in rapid serial visual presentation, accuracy for reporting the second target is markedly diminished-an attentional blink (AB). The AB has become a well-studied phenomenon, and multiple different versions of the AB are currently in use. However, little is known about the stability of individual performance on the AB. The current study examined the reliability of two different versions of the AB task (a task-switch and no-task-switch version) within session, and over the period of 7-10 days, in order to examine performance stability. In addition to testing the reliability, we also examined the relationship between both versions of our AB tasks. Both versions of the AB were shown to be reliable within session, and over time, suggesting that performance is quite stable on this task. In addition, performance on the two different AB tasks was significantly correlated within and across sessions, suggesting that the AB phenomenon is being accurately captured by versions of the AB that include a task-switch. These findings are important given the recent interest in individual differences in performance on the AB.

How reliable is the attentional blink? Examining the relationships within and between attentional blink tasks over time.

When the second of two targets (T2) is presented in close temporal proximity (within 200-500 ms) to the first (T1), the accuracy for reporting T2 is reduced relative to when the targets are separated by longer durations; this effect is known as the attentional blink (AB). Two recent studies have shown that individual differences in the magnitudes of the AB are stable both within a single testing session and over time. While one study found a large positive correlation between AB magnitudes when there was an attentional-set/task switch between T1 and T2 and when there was not, the other study found no relationship between the switch and no-switch paradigms. The present study was conducted to clarify this discrepancy by examining the reliability of, and relationships among, individual differences in AB performance on five different versions of the standard dual-target RSVP paradigm, three of which involved an attentional-set/task switch between T1 and T2, and two of which did not. Participants completed all five paradigms, and then returned 7-10 days later to again complete the same paradigms. The performance on all five versions was reliable both within and across testing sessions, demonstrating again that individual differences in AB performance are stable over time. In addition, all five AB versions were significantly intercorrelated, although the strengths of the relationships differed depending on the extent to which the T1 and T2 attentional sets/tasks overlapped. These findings provide evidence that multiple distinct dual-target RSVP tasks do share underlying variability, providing support for the use of different versions of the paradigm in the literature.

Individual differences within and across attentional blink tasks revisited.

Global/local stimuli have been used to estimate global processing biases in individuals and groups, as well as in response to various manipulations. Throughout the literature, multiple different versions of global/local stimuli have been used, such as traditional hierarchical letters and numbers (i.e., Navon letters), abstract hierarchical shapes, and high- and low-spatial-frequency gratings and faces. However, currently it is unclear how reliable or stable performance is on these measures within individuals over time, and whether these seemingly different measures are tapping into the same underlying process. As such, the purpose of the present study was to examine the stability of individual performance on three distinct global/local measures over time and to examine the relationships among the measures. In two studies, we examined the reliability of the biases within, and the relationships among, standard Navon letters in a traditional interference task, hierarchical shapes in a forced choice task, and superimposed high- and low-pass spatial frequency faces in a forced choice task. In both studies, participants completed all three of the tasks, and then returned 7-10 days later to again complete the same tasks. The degree of global/local bias within an individual was found to be highly reliable in the hierarchical shape task and the spatial frequency face task, but less reliable in the traditional Navon letter task. Interestingly, in both studies we found that none of the three measures of global bias were related to each other. Therefore, while these measures do appear to be reliable over time, they may be tapping into distinct aspects of global/local processing.

Investigating the stability of and relationships among global/local processing measures.

Visual stimuli can be perceived at a broad, "global" level, or at a more focused, "local" level. While research has shown that many individuals demonstrate a preference for global information, there are large individual differences in the degree of global/local bias, such that some individuals show a large global bias, some show a large local bias, and others show no bias. The main purpose of the current study was to examine whether these dispositional differences in global/local bias could be altered through various manipulations of high/low spatial frequency. Through 5 experiments, we examined various measures of dispositional global/local bias and whether performance on these measures could be altered by manipulating previous exposure to high or low spatial frequency information (with high/low spatial frequency faces, gratings, and Navon letters). Ultimately, there was little evidence of change from pre-to-post manipulation on the dispositional measures, and dispositional global/local bias was highly reliable pre- to post-manipulation. The results provide evidence that individual differences in global/local bias or preference are relatively resistant to exposure to spatial frequency information, and suggest that the processing mechanisms underlying high/low spatial frequency use and global/local bias may be more independent than previously thought.

Lost in the forest, stuck in the trees: dispositional global/local bias is resistant to exposure to high and low spatial frequencies.

When the second of two targets (T2) is presented temporally close to the first target (T1) in a rapid serial visual presentation stream, accuracy to identify T2 is markedly reduced-an attentional blink (AB). While most individuals show an AB, Dale and Arnell (Atten Percept Psychophys 72(3):602-606, 2010) demonstrated that individual differences in dispositional attentional focus predicted AB performance, such that individuals who showed a natural bias toward the global level of Navon letter stimuli were less susceptible to the AB and showed a smaller AB effect. For the current study, we extended the findings of Dale and Arnell (Atten Percept Psychophys 72(3):602-606, 2010) through two experiments. In Experiment 1, we examined the relationship between dispositional global/local bias and the AB using a highly reliable hierarchical shape task measure. In Experiment 2, we examined whether three distinct global/local measures could predict AB performance. In both experiments, performance on the global/local tasks predicted subsequent AB performance, such that individuals with a greater preference for the global information showed a reduced AB. This supports previous findings, as well as recent models which discuss the role of attentional breadth in selective attention.

Gillian Dale

Department of Psychology

University of Wisconsin-Madison

Methods to Test Visual Attention Online