| |
Abstract:
When two targets (T1 and T2) are processed in close temporal
proximity (100-500ms), the processing of the second target is often
disrupted (the attentional blink). T2 accuracy is usually preserved
at short temporal lags, disrupted at intermediate lags, and
unaffected at long lags. This non-monotonic pattern of interference
suggests limitations in post-perceptual resources needed for
processing each target. The preserved performance at short lags
(lag 1 sparing) has typically been observed at a T1-T2 stimulus
onset asynchrony (SOA) of 100 ms, and it has been attributed to the
slow closing of an attentional gate. By this view, T2 gains access
to later processing stages because of its temporal proximity with
T1. However, Visser et al. (1999) reported an absence of lag 1
sparing when T1 and T2 appeared in different locations. They
suggested that a spatially-specific attentional gate opens at the
location of T1, and that T2 cannot be processed until a new gate
opens at its location. To explore this hypothesis, we polled T2
accuracy at SOAs shorter than 100ms and varied the timing of the T1
mask. We found that T2 performance was preserved at SOAs shorter
than 100 ms. Furthermore, when perceptual interference between the
T1 mask and T2 was eliminated, typical lag 1 sparing effects were
reinstated. Lag1 sparing is better predicted by temporal rather
than spatial contiguity.
|
