Gatecrashing the visual cocktail party: How visual and semantic similarity modulate the own name benefit in the attentional blink

The “visual cocktail party effect” refers to superior report of a participant’s own name, under conditions of inattention. An early selection account suggests this advantage stems from enhanced visual processing. A late selection account suggests the advantage occurs when semantic information allowing identification as one’s own name is retrieved. In the context of inattentional blindness (IB), Mack and Rock showed that the advantage does not generalise to a minor modification of a participant’s own name, despite extensive visual similarity, supporting the late selection account. This study applied the name modification manipulation in the context of the attentional blink (AB). Participants were presented with rapid streams of names and identified a white target name, while also reporting the presence of one of two possible probes. The probe names appeared either close (the third item following the target: Lag 3) or far in time from the target (the eighth item following the target: Lag 8). The results revealed a robust AB; reports of the probe were reduced at Lag 3 relative to Lag 8. The AB was also greatly reduced for the own name compared to another name—a visual cocktail party effect. In contrast to the findings of Mack and Rock for IB, the reduced AB extended to the modified own name. The results suggest different loci for the visual cocktail party effect in the AB (word recognition) compared to IB (semantic processing).

Inhibition of Return Effect Reversed by Perceptual Load

Perceptual load theory was considered to resolve the conflict between early and late selection models. We investigated whether perceptual load work well using IOR paradigm. The results showed that IOR disappeared in different level of perceptual load. The results indicated the onset time of IOR was influenced by perceptual load, indicating that perceptual load hinder the spatiotemporal correspondence between the cue and target integration.

The Dissociation between Early and Late Selection in Older Adults

Older adults exhibit a reduced ability to ignore task-irrelevant stimuli; however, it remains to be determined where along the information processing stream the most salient age-associated changes occur. In the current study, ERPs provided an opportunity to determine whether age-related differences in processing task-irrelevant stimuli were uniform across information processing stages or disproportionately affected either early or late selection. ERPs were measured in young and old adults during a color-selective attention task in which participants responded to target letters in a specified color (attend condition) while ignoring letters in a different color (ignore condition). Old participants were matched to two groups of young participants on the basis of neuropsychological test performance: one using age-appropriate norms and the other using test scores not adjusted for age. There were no age-associated differences in the magnitude of early selection (attend–ignore), as indexed by the size of the anterior selection positivity and posterior selection negativity. During late selection, as indexed by P3b amplitude, both groups of young participants generated neural responses to target letters under the attend versus ignore conditions that were highly differentiated. In striking contrast, old participants generated a P3b to target letters with no reliable differences between conditions. Individuals who were slow to initiate early selection appeared to be less successful at executing late selection. Despite relative preservation of the operations of early selection, processing delays may lead older participants to allocate excessive resources to task-irrelevant stimuli during late selection.