| |
Abstract:
Hemodynamic neuroimaging methods are used extensively to
study brain function. However, the quantitative relationship
between hemodynamic phenomena and neuronal activity is still
debated. Our previous studies indicate that non-invasive optical
imaging may be an appropriate tool for investigating this
relationship because it can provide simultaneously localized
indices of both neuronal responses (related to changes in light
scattering) and hemodynamic phenomena (related to changes in light
absorption). In the present study we presented six subjects with a
visual paradigm in which the frequency of stimulation (reversals of
a vertical grid) was 1, 2, 5, or 10 Hz in different blocks. The
fast (neuronal) response was quantified as the change in the
near-infrared photon transit time through medial occipital areas at
a latency of 60-100 ms after each grid reversal. The slow
(hemodynamic) response was quantified as the average change in the
amount of light transiting through the same area over the entire
stimulation period (16 s). The amplitude of the fast response was
steady for stimulation frequencies up to 5 Hz, but declined at 10
Hz. The slow response increased linearly up to 5 Hz and then
declined. The slow response was proportional to the fast response
multiplied by the stimulation frequency (r=.99). The data support
the hypothesis that the hemodynamic phenomenon studied with
neuroimaging methods is linearly related to amplitude of the
neuronal response.
|
