Recent studies have linked variability in near-threshold stimulus detection to fluctuations in the prestimulus EEG alpha power (α, ∼8-12 Hz). Typically, these studies rely on hit rate as a measure of detection performance and show that detection is enhanced when α power is low compared to when it is high. However, hit rates are determined by both sensitivity to the stimulus and the placement of the response criterion. Here, we investigated the relationships between prestimulus α power and variability in these two measures on a single-trial basis. We confirm earlier reports that detection is inversely related to power in the individual α-frequency band. However, our results show a stronger relationship between α power and response criterion than with sensitivity. Higher α power was related to a more conservative response criterion (i.e., more "no" responses). A response criterion that varies depending on α power might help to optimize performance in an excited state and protect against false positives in a relatively disengaged state.

No abstract

Recent theories about human brain function emphasize the need for imaging methods that allow the study of dynamic interactions among different structures. In this paper, we report on a new technique, based on the measurement of parameters of migration of near-infrared photons, that yields functional images of the human occipital cortex, combining a spatial resolution of 0.5 cm and a temporal resolution of 50 ms. This technique appears to be suitable for studying the dynamics of cortical activation.

No abstract

Localized attentional interference (LAI) occurs when attending to a visual object degrades processing of nearby objects. Competitive interaction accounts of LAI explain the phenomenon as the result of competition among objects for representation in extrastriate cortex. Here, we examined the N2pc component of the event-related potential (ERP) as a likely neural correlate of LAI. In Experiment 1, participants responded to the orientation of a target while ignoring a nearby decoy. At small target-decoy separations, N2pc amplitude was attenuated whereas the amplitude of a later, positive component (Ptc) was potentiated. Experiment 2 ruled out sensory explanations of these effects. The N2pc results are consistent with the idea that spatially mediated competition for representation in extrastriate cortex degrades target selection. Moreover, the Ptc may reflect a bias signal needed to resolve the competition at smaller target-decoy separations.