Low-level masking in the attentional blink.

Abstract

If two targets (Tl and T2) are to be identified among other stimuli displayed in rapid serial visual presentation (RSVP), correct identification of Tl can produce an attentional blink (AB) lasting several hundred milliseconds, during which detection of T2 is impaired. Experiment 1 confirmed that omission of the item directly following Tl (the +1 item) reduces the AB (J. E. Raymond, K. L. Shapiro, & K. M. Arnell, 1992). The next 3 experiments varied the spatial and temporal relationships between Tl and the +1 item to study how masking of Tl affects the AB deficit. Perception of Tl was impaired by pattern masking arising from temporal integration or superimposition of Tl and die +1 item; it was also impaired by metacontrast masking. We conclude that masking affects the AB indirectly by degrading Tl thereby increasing the duration of Tl processing. A 2-stage model proposed by M. M. Chun and M. C. Potter (1995) is supported. Sequential visual stimuli-as might be generated from objects in motion or from eye movements in everyday viewing-can pose a problem for human information processing. If new stimuli arrive at a rate that exceeds the visual system's temporal capabilities, some may fail to be perceived. With stimuli shown in a rapid serial visual presentation (RSVP), this failure has been attributed to attentional processes: If two targets are to be identified among distractors in an RSVP stream, correct identification of the first target (Tl) may produce an attentional blink (AB) lasting several hundred milliseconds, during which detection of the second target (T2) is impaired (Raymond, Shapiro, & Ar- nell, 1992). In the present work we show that, in addition to high-level attentional processes, low-level visual processes-notably masking by pattern and metacontrast masking-contribute to the AB deficit. Such low-level processes provide an explanatory basis for several phenomena that cannot be accounted for directly in current theories. The effect of low-level processes can be studied by examining the role played by the distractor that comes directly after Tl in the RSVP stream.Presence of a distractor item in the frame directly following Tl has been regarded as essential for obtaining an AB deficit. If that item (known as the +1 item) is replaced by a

The Role of the Locus Coeruleus in Mediating the Attentional Blink: A Neurocomputational Theory.

“…It is generally thought that this effect is mediated by backward masking of T1 by the Lag 1 item (e.g., Brehaut, Enns, & Di Lollo, 1999;Seiffert & Di Lollo, 1997). Our computational model does not incorporate specific mechanisms for simulating backward masking and, hence, does not directly address the effect of T1 masking on T2 performance.…”

Spatiotemporal Segregation in Visual Search: Evidence From Parietal Lesions.

“…Some studies have manipulated the spatial locations of the items in an RSVP stream and demonstrated that the attentional blink extends across space. For instance, Visser, Zuvic, Bischof, and Di Lollo (1999; see also Breitmeyer, Ehrenstein, Pritchard, Hiscock, & Crisan, 1999;Seiffert & Di Lollo, 1997;Shih, 2000) found that the identification of a second target is also impaired when it is presented to the left or right of the first target. They argued that attention cannot be switched to a new location while the system is processing the first target.…”

The beneficial effects of additional task load, positive affect, and instruction on the attentional blink.

“…In this sense, we could regard T1 and T2 as two communicating vessels; the more resources required for the one, the fewer are left for the other. Indeed, various studies have confirmed a negative correlation between behavioral indices or neural correlates of T1 and T2 processing (Chun & Potter, 1995, Table 1;Hommel et al, 2003;Seiffert & Di Lollo, 1997;Shore et al, 2001; see also Kahneman, 1973, p. 151). However, note that in most of our experiments, if anything, T1 detection improved when observers were associating freely (Olivers & Nieuwenhuis, 2005), listening to music (Olivers & Nieuwenhuis, 2005), doing another task (Experiment 1), or being presented with positive stimuli (Experiment 2).…”

“…Various explanations of the attentional blink have been proposed (Chun & Potter, 1995;Duncan, Ward, & Shapiro, 1994;Giesbrecht & Di Lollo, 1998;Jolicoeur & Dell'Acqua, 1998;Nieuwenhuis, Gilzenrat, Holmes, & Cohen, 2005;Raymond et al, 1992;Seiffert & Di Lollo, 1997;Shapiro & Raymond, 1994;Vogel, Luck, & Shapiro, 1998), but they all stress the limitedcapacity nature of attentional processing as the main determinant. For example, Chun and Potter (1995) proposed a two-stage theory.…”

“…Various studies have shown that the fewer the processing resources required for (or assigned to) T1, the weaker the attentional blink for T2 (see Chun & Potter, 1995, Table 1; Seiffert & Di Lollo, 1997;Shore, McLaughlin, & Klein, 2001; see also Hommel, Schmitz, Shapiro, & Schnitzler, 2003). However, we found that detection of T1 did not deteriorate (if anything it improved slightly).…”