On the function of object cells in the claustrum—key components in information processing in the visual system?

A recent paper by Jankowski and O’Mara (2015) throws new light on the function of the claustrum. These authors took recordings from 874 claustral neurons in unanesthetized behaving rats that showed responses probably anchored to visual stimuli. These fell into three categories—“place” neurons (38) that responded to the location of the rat, “boundary” neurons (23) that responded to boundaries closing the rat’s environment, and “object” neurons (48) that fired only in response to the existence of a specific object in the environment. All three types of response showed with good within-session stability. Of particular interest in the present context are the object cells. These code the position in space of an object in the environment and not its particular properties (e.g., shape, texture, color). They also follow an object when it moves in the environment. Some object neurons can follow multiple objects (up to 3). Object neurons are also found in the subiculum, and in the anterior entorhinal, perirhinal, and anterior cingulate cortices with which areas the claustrum has extensive connections. The authors suggest that this system processes dynamic information about spatial features of the environment relating to the location in space of the organism, and the presence of important features in the environment such as boundaries and objects. The claustrum then supplies this information to the wider brain systems in the hippocampus and related cortices so as to facilitate the moment-to-moment control of behavior. The question this opinion piece addresses is how does this new data relate to a current hypothesis as to the function of the claustrum first put forward by Crick and Koch (2005) and elaborated by Smythies et al. (2012, 2014)? This hypothesis suggests that the claustrum integrates the sensorimotor mechanisms that control behavior via competitive “winner-takes-all” synchronized gamma oscillations in the cortico-claustral circuits by the Pearson system. This system states that, if two cortical areas A and B connect by synchronized oscillations, and, if A connects to claustral area C and B connects to claustral area D, then C, and D will be connected by synchronized oscillations. However, if A and B are not so connected, then C and D will not be connected either. One “low level” feature of our original hypothesis is that the “binding” of disparate color, shape and movement sensations in the perception of a unitary perceptual object may mediated by this synchronization of gamma oscillations. Another “high level” feature of this hypothesis, also using competitive synchronized oscillations, is that circuits linking the claustrum with higher cortex mediate voluntary decisions involving the selection of particular behaviors in response to complex sensory and multisensory inputs. Four recent papers are relevant to this hypothesis.