Datensatz

Zusammenfassung

An increasing body of literature provides compelling evidence that sensory convergence not
only occurs in higher association areas, but also in lower sensory regions and even in primary
sensory cortices. To scrutinize these early cross-modal interactions, we use the macaque auditory
cortex as model and employ combinations of high-resolution functional imaging (fMRI)
and electrophysiological recordings. Using function imaging in alert and anaesthetized animals,
we reported that (only) caudal auditory fields are susceptible to cross-modal modulation:
The fMRI-BOLD response in these regions was enhanced when auditory stimuli were complemented
by simultaneous visual or touch stimulation [1,2]. To investigate the neuronal basis
of this cross-modal enhancement, we recorded the activity of local field potentials and single
units in alert animals watching complex audio-visual scenes. Our results show the following:
Visual stimuli by themselves, on average, do not drive auditory neurons, but cause responses
in low frequency LFPs. Combining visual and auditory stimuli leads to enhanced responses
in the low frequency LFP, but to a reduction of firing rates. This audio-visual interaction was
significant at the population level, and for about 10 of the neurons when tested individually.
The interaction occurs only for well-timed visual stimuli, is strongest when the visual stimulus
leads the auditory stimulus by 20–80msec, but is independent of the image structure in the visual
stimulus. Similar visual modulation was found in the auditory core and belt. Our findings
point to a very basic, stimulus unspecific visual input to auditory cortex and clearly support
the notion that early sensory cortices are susceptible to cross-modal interactions. Especially,
the finding that visual stimuli modulate the firing rates of individual neurons in auditory cortex
suggests that the messages transmitted from these regions to higher processing stages do not
only reflect acoustical stimuli but are also dependent on their visual context.