There is increasing evidence of separate auditory-cortex pathways for object and spatial features 
, analogous to the parallel “what”
visual pathways 
. Given the existing knowledge of crossmodal connections 
, the auditory “what”
pathways may separately interact with their visual counterparts at multiple levels 
. However, the exact intersections where the auditory and visual dual pathways meet to govern processing still remain unknown, especially when it comes to attentional modulations.
In the spatial domain, attention to auditory or visual locations activates largely overlapping parietal networks 
(although some evidence for modality-specific nodes exists 
). Audiospatial attention systems are often considered subsidiary to visuospatial processes 
. Indeed, auditory stimuli are more easily mislocalized toward concurrent but spatially incongruent visual events than vice versa
. However, crossmodal influences in the opposite direction occur as well 
: Audiospatial attention may govern visual orienting to out-of-view stimuli 
and improve detection of unexpected visual targets in expected locations of auditory targets 
. The posterior audiospatial processing stream may also play a critical role in guiding motor and visuomotor processes 
Object-centered multisensory attention is less clearly understood. A recent EEG study 
suggested that attentional control over auditory and visual “what”
streams is predominantly modality specific. However, sound-object perception can certainly be affected by crossmodal information. For example, visual attention to speakers' lips can modulate perception of ambiguous auditory speech objects 
, and even alter the percepts 
. Conversely, sounds may affect perception of visual objects 
and help select relevant events in an environment containing multiple competing visual objects 
. Recent studies also suggest that conflicting auditory objects may modulate the spread and capture of visual object-related attention across multisensory objects 
, and that attending to either a visual or an auditory object results in a co-activation of the attended stimulus representation in the other modality 
. Further studies are, thus, needed to elucidate multisensory aspects of spatial vs
. object-specific attention.
Attention is reflected by modulations in neuronal oscillations, non-invasively measurable with magnetoencephalography (MEG) and EEG. Previous studies suggest that the degree of oscillatory synchronization may tell us whether a spatially confined, local neuronal group is processing an attended stimulus effectively 
. Different aspects of attentional modulations of brain activity may, however, occur in distinct frequency bands. Neurophysiological studies in the macaque visual cortex, for example, suggest that neurons activated by an attended stimulus show increased synchronization at higher-frequency gamma band (~35–90 Hz) and decreased synchronization at lower frequency bands (<17 Hz) 
. Analogous effects have been well documented also in human MEG and EEG studies. That is, increased attentional processing in areas representing task-relevant stimuli has been shown to increase gamma power in human visual 
, auditory 
, and somatosensory 
visual cortices, while increased synchronization at the lower frequency bands, particularly at the alpha range (~7–13 Hz), has been associated with disengagement of a network representing task-irrelevant stimulus features 
Alpha rhythms are a ubiquitous oscillatory phenomenon whose modulations by subjects' alertness and attentional state may be readily observed in the raw MEG/EEG traces even without signal analysis tools. Alpha oscillations increase, for instance, during drowsiness and limited visual input and, conversely, decrease during visual stimulation and tasks 
, which has led to the prevailing interpretation that enhanced alpha activity reflects “idling” 
or “active inhibition” 
. Consistent with this view, when visual attention is strongly focused to one location of visual field, alpha activity may significantly increase in retinotopic visual-cortex areas representing other (i.e.
, task-irrelevant) aspects of the visual field, possibly reflecting active inhibition of activity in the underlying populations 
. Such alpha inhibition effects have been shown to correlate with the ability to ignore irrelevant visual stimuli 
. Not surprisingly, parieto-occipital alpha also increases when auditory 
or somatosensory 
instead of visual stimuli are attended. Task-related alpha modulations might, thus, help measure associations between auditory and visual attention networks. Here, we used MEG to study how object vs.
spatial auditory attention affects cortical alpha oscillations generated in the absence of visual stimuli or tasks.