Abstract
How does temporal selection work, and along what dimensions does it vary from one instance to the next? We explored these questions using a phenomenon in which temporal selection goes awry. In the attentional blink, subjects fail to report the second of a pair of targets (T1 and T2) when they are presented at stimulus onset asynchronies (SOAs) of roughly 200 to 500 ms. We directly tested the properties of temporal selection during the blink by analyzing distractor intrusions at a fast rate of item presentation. Our analysis shows that attentional selection is (a) suppressed, (b) delayed, and (c) diffused in time during the attentional blink. These effects are dissociated by their time course: The measure of each effect returns to the baseline value at a different SOA. Our results constrain theories of the attentional blink and indicate that temporal selection varies along at least three dissociable dimensions: efficacy, latency, and precision.
Introduction
Human attention is limited in time. When identifying two targets (T1 and T2) in a rapid stream of information, subjects often miss the second target if it appears within 200–500 milliseconds of the first. This phenomenon, known as the Attentional Blink (AB), has traditionally been viewed as a binary "hit or miss" failure. However, this study argues that such a coarse metric obscures the underlying mechanics of temporal selection. By analyzing the specific errors made during the blink, we can determine if the attentional gate simply fails to open (suppression), opens too late (delay), or loses its temporal focus (diffusion).
Methods
Researchers employed a Rapid Serial Visual Presentation (RSVP) task where twelve participants viewed a stream of 26 letters presented at a rate of 12 items per second (25ms per item). Two targets were cued by white rings. Crucially, the study analyzed distractor intrusions—the specific letters reported when the target was missed. These intrusions allowed for the mathematical decomposition of attentional selection into three dimensions:
- Efficacy (Suppression): The probability of reporting any item near the target.
- Latency (Delay): The temporal "center of mass" of the reports.
- Precision (Diffusion): The temporal spread or variance of the reports.
Key Results: The Triple Threat
The analysis revealed that the attentional blink is not a single failure but a complex breakdown across three dissociable dimensions.
Suppression and Diffusion
During the peak of the blink (166–417ms), there was a significant drop in the probability of reporting an item from the target window, indicating suppression. Simultaneously, the variance of reports increased, showing that selection became diffused or "blurry" in time.
Delay
Selection was also significantly delayed. At an SOA of 417ms, selection was shifted by approximately 75ms. Interestingly, a subtle delay of about 30ms persisted even at the longest interval tested (833ms), suggesting a long-lasting "hangover" effect of the blink.
Temporal Dissociation and Lag-1 Sparing
The most significant finding was that these three deficits recover at different rates. Diffusion returns to baseline first (around 400ms), followed by suppression (by 500ms), while delay is the most persistent, lasting beyond 800ms. Notably, at the shortest interval (83ms), efficacy remained high, confirming the "Lag-1 Sparing" phenomenon where T2 escapes the blink by appearing immediately after T1.
Conclusion
The attentional blink involves a multifaceted degradation of temporal selection. Because the dimensions of suppression, delay, and diffusion follow different time courses, they cannot be explained by a single-process model. These results indicate that temporal selection is a complex mechanism varying along at least three dissociable dimensions, providing new constraints for theories of human attention.
Summary updated