The dynamics of buffered and triggered selection from rapid serial visual presentation (RSVP) streams.

The temporal dynamics of selective attention are reflected by distractor intrusions

“…This account predicts that when the probability of selection failure increases by introducing uncertainty about target location in the two-stream condition, this should generally increase both pre-target and post-target distractor intrusions in Experiment 1, and have similar effects on the probability of pre-cue and post-cue digit reports in Experiment 2. In contrast, our results are entirely compatible with the attentional episode account 9 , 11 . According to this account, there is no qualitative difference between the processes that result in correct reports and in distractor intrusions.…”

“…In both experiments, we assumed that increasing the number of streams from one to two slowed attentional engagement 9 , 12 , 22 , 23 , and that the differential effects on pre-target and post-target reports were associated with this delay. However, it is possible that this manipulation may have had additional effects that could also explain our results.…”

“…If intrusions are linked to the temporal distribution of attentional enhancement, a manipulation that delays attentional engagement should increase post-target intrusions but reduce pre-target intrusions (as they shall benefit even less from attentional enhancement). Recently, Ludwici and Holcombe 9 used a similar manipulation and found evidence for such a pattern. They concluded that perceptual representations of pre-target items persist for a short amount of time and that these buffered representations may be selected for report when attentional engagement is fast enough to promote their processing.…”

The number of expected targets modulates access to working memory: A new unified account of lag-1 sparing and distractor intrusions.

“…Perhaps unsurprisingly, task performance was generally better in blocks with a single central RSVP stream relative to blocks with two lateral streams, with higher T1 accuracy and lower guess rates for single-stream blocks. As target location was certain in single-stream blocks, this is compatible with the notion that focused spatial attention enhances early perceptual processing (Luck et al, 1997) and results in faster target detection and consequently in earlier attentional episodes (Foster et al, 2020; Zivony & Eimer, 2021b, Experiment 3; see also Ludowici & Holcombe, 2021). It is also notable that only T1 reports but not reports of the post-T1 item showed benefits in single-stream blocks.…”

“…Accuracy is high (about 80–90%), when this posttarget distractor does not match the response category of the target (e.g., when a digit target is followed by a letter), but drops precipitously (to 40–50%; e.g., Zivony & Eimer, 2021b) when this both items come from the same category (e.g., a digit target followed by a digit distractor). In the latter case, observers will often erroneously report the identity of the posttarget distractor instead of the target (e.g., Botella et al, 2001; Goodbourn et al, 2016; Ludowici, & Holcombe, 2021; Vul et al, 2008).…”