What processes are disrupted during the attentional blink? An integrative review of event-related potential research

Abstract

Reporting the second of two targets is impaired when these appear in close succession, a phenomenon known as the attentional blink (AB). Despite decades of research, what mechanisms are affected by the AB remains unclear. Specifically, two central issues remain open: Does the AB disrupt attentional processes or reflect a structural limitation in working memory encoding? Does it disrupt perceptual processing or only post-perceptual processes? We address these questions by reviewing event-related potentials (ERP) studies of the AB. The findings reveal that the core influence of the AB is by disrupting attentional engagement (indexed by N2pc). As a consequence, while early processing (indexed by P1\N1) is spared, semantic processing (indexed by N400) and working memory (WM) encoding (indexed by P3b) are compromised: minor disruptions to attentional engagement weaken but do not eliminate semantic processing, whereas they prevent encoding in WM. Thus, semantic processing can survive the blink, whereas encoding in WM does not. To accommodate these conclusions, we suggest a Disrupted Engagement and Perception (DEaP) account of the attentional blink.

Graded Visual Consciousness During the Attentional Blink

“…Thus, differences in visibility ratings cannot be attributed to differences in low-level visual processing of T2 stimuli. This result is consistent with overall evidence indicating that activity during the P1 time range is not modulated by the attentional blink (see Zivony & Lamy, 2020). In the consciousness literature as a whole, there also exists comparatively little evidence for differences in the P1 time range, and by now, there seems to be a growing consensus that the P1 is not a marker of conscious awareness (Förster et al, 2020;Koivisto & Revonsuo, 2010).…”

“…Already in the N1/N170 component, significant gradations of activity depending on the reported visibility were found. This effect stands in contrast to the attentional blink literature in general (when comparing "seen" and "unseen" trials) which overall indicates no differences during this time range (see Zivony & Lamy, 2020). In Sergent et al (2005), also no differences were observed in the N1 component.…”

“…Differences between detected and undetected stimuli during this time period have frequently been reported in the attentional blink literature (Koivisto & Revonsuo, 2008;Sergent et al, 2005;Zivony & Lamy, 2020; yet, this has not always been found, see e.g., Batterink et al, 2012). In terms of functional significance, modulations during this time range are assumed to reflect differences in attentional engagement (Zivony & Lamy, 2020). The graded pattern observed here is consistent with the one reported by Sergent et al (2005) where graded differences were observed for the comparison of four visibility levels (which were obtained by separating the used 21-point scale into four categories).…”

“…Significant graded differences for visibility levels were also found in the N2 time range. Differences between detected and undetected stimuli during this time period have frequently been reported in the attentional blink literature (Koivisto & Revonsuo, 2008;Sergent et al, 2005;Zivony & Lamy, 2020; yet, this has not always been found, see e.g., Batterink et al, 2012). In terms of functional significance, modulations during this time range are assumed to reflect differences in attentional engagement (Zivony & Lamy, 2020).…”

“…The attentional blink describes the phenomenon that in a rapid serial visual presentation (RSVP) task, the detection of the second (T2) of two target stimuli presented in a stream of distractor stimuli is often impaired if it follows the first target stimulus (T1) within a period of 200 to 500 ms as compared to a longer (or shorter) interval between both targets (Raymond et al, 1992). As yet, the mechanisms behind this phenomenon are not fully resolved but in general terms explained by a disruption of attentional processes and/or of encoding in working memory due to the ongoing processing of T1 (for a recent review, see Zivony & Lamy, 2020). Often in investigations of the attentional blink only the contrast between trials in which T2 was "seen" versus "not seen" (or "detected" versus "not detected") was considered, which implicitly contains the assumption of an all-or-none pattern of conscious perception.…”