An intriguing aspect of the data in Clark et al. is the observation
that interpeduncular nucleus (IPN) lesions increase (by about a factor 2)
the amount by which thalamic HD cells anticipate future head direction.
Integration of such anticipatory time interval (ATI) data with modeling
studies has been a useful tool in dissecting the HD system circuitry. An
important question in this respect is whether anticipation in the HD
system results from the same mechanisms responsible for updating the
system in response to movement, or if the two are distinct.
Clark et al. find a dissociation of IPN contributions to the update
mechanism (decreased accuracy if lesioned), and anticipation (increased if
lesioned). This suggests not only that the putative motor contribution of
IPN cannot be the source of HD anticipation, but moreover, that in the
intact animal, such input serves to counteract anticipation originating
from a different source. This interpretation is further supported by a
previous report from the same group (Bassett and Taube 2005) showing that
when animals are rotated passively, thalamic HD cells anticipate more than
during active movement. Taken together, these data appear inconsistent
with the proposal that HD anticipation might derive from motor efference
copy, and instead support its generation from vestibular or network
properties.
However, anticipatory time interval (ATI) data needs to be treated
with caution. In particular, van der Meer et al. (2007) found a
relationship between the dynamics of head movements and observed ATIs,
where rapid, abrupt and high-frequency head movements are associated with
little anticipation, and smooth movement with high ATIs. In order to
increase confidence in the above interpretation of the Clark et al. ATI
results, it would be helpful to know if there are any differences in head
movements associated with IPN lesions that may account for the observed
effect on anticipation.