A central role for the retrosplenial cortex in de novo environmental learning

Multivoxel Pattern Analysis Reveals 3D Place Information in the Human Hippocampus

The spatial world is three dimensional (3D) and humans and other animals move both horizontally and vertically within it. Extant neuroscientific studies have typically investigated spatial navigation on a horizontal 2D plane, leaving much unknown about how 3D spatial information is represented in the brain. Specifically, horizontal and vertical information may be encoded in the same or different neural structures with equal or unequal sensitivity. Here, we investigated these possibilities using fMRI while participants were passively moved within a 3D lattice structure as if riding a rollercoaster. Multivoxel pattern analysis was used to test for the existence of information relating to where and in which direction participants were heading in this virtual environment. Behaviorally, participants had similarly accurate memory for vertical and horizontal locations and the right anterior hippocampus (HC) expressed place information that was sensitive to changes along both horizontal and vertical axes. This is suggestive of isotropic 3D place encoding. In contrast, participants indicated their heading direction faster and more accurately when they were heading in a tilted-up or tilted-down direction. This direction information was expressed in the right retrosplenial cortex and posterior HC and was only sensitive to vertical pitch, which could reflect the importance of the vertical (gravity) axis as a reference frame. Overall, our findings extend previous knowledge of how we represent the spatial world and navigate within it by taking into account the important third dimension.

SIGNIFICANCE STATEMENT The spatial world is 3D. We can move horizontally across surfaces, but also vertically, going up slopes or stairs. Little is known about how the brain supports representations of 3D space. A key question is whether horizontal and vertical information is equally well represented. Here, we measured fMRI response patterns while participants moved within a virtual 3D environment and found that the anterior hippocampus (HC) expressed location information that was sensitive to the vertical and horizontal axes. In contrast, information about heading direction, found in retrosplenial cortex and posterior HC, favored the vertical axis, perhaps due to gravity effects. These findings provide new insights into how we represent our spatial 3D world and navigate within it.

Making Sense of Real-World Scenes

To interact with the world, we have to make sense of the continuous sensory input conveying information about our environment. A recent surge of studies has investigated the processes enabling scene understanding, using increasingly complex stimuli and sophisticated analyses to highlight the visual features and brain regions involved. However, there are two major challenges to producing a comprehensive framework for scene understanding. First, scene perception is highly dynamic, subserving multiple behavioral goals. Second, a multitude of different visual properties co-occur across scenes and may be correlated or independent. We synthesize the recent literature and argue that for a complete view of scene understanding, it is necessary to account for both differing observer goals and the contribution of diverse scene properties.

Optimal cue combination and landmark-stability learning in the head direction system

Maintaining a sense of direction requires combining information from static environmental landmarks with dynamic information about self-motion. This is accomplished by the head direction system, whose neurons - head direction cells - encode specific head directions. When the brain integrates information in sensory domains, this process is almost always 'optimal' - that is, inputs are weighted according to their reliability. Evidence suggests cue combination by head direction cells may also be optimal. The simplicity of the head direction signal, together with the detailed knowledge we have about the anatomy and physiology of the underlying circuit, therefore makes this system a tractable model with which to discover how optimal cue combination occurs at a neural level. In the head direction system, cue interactions are thought to occur on an attractor network of interacting head direction neurons, but attractor dynamics predict a winner-take-all decision between cues, rather than optimal combination. However, optimal cue combination in an attractor could be achieved via plasticity in the feedforward connections from external sensory cues (i.e. the landmarks) onto the ring attractor. Short-term plasticity would allow rapid re-weighting that adjusts the final state of the network in accordance with cue reliability (reflected in the connection strengths), while longer term plasticity would allow long-term learning about this reliability. Although these principles were derived to model the head direction system, they could potentially serve to explain optimal cue combination in other sensory systems more generally.

The occipital place area represents first-person perspective motion information through scenes

Neuroimaging studies have identified multiple scene-selective regions in human cortex, but the precise role each region plays in scene processing is not yet clear. It was recently hypothesized that two regions, the occipital place area (OPA) and the retrosplenial complex (RSC), play a direct role in navigation, while a third region, the parahippocampal place area (PPA), does not. Some evidence suggests a further division of labor even among regions involved in navigation: While RSC is thought to support navigation through the broader environment, OPA may be involved in navigation through the immediately visible environment, although this role for OPA has never been tested. Here we predict that OPA represents first-person perspective motion information through scenes, a critical cue for such "visually-guided navigation", consistent with the hypothesized role for OPA. Response magnitudes were measured in OPA (as well as RSC and PPA) to i) video clips of first-person perspective motion through scenes ("Dynamic Scenes"), and ii) static images taken from these same movies, rearranged such that first-person perspective motion could not be inferred ("Static Scenes"). As predicted, OPA responded significantly more to the Dynamic than Static Scenes, relative to both RSC and PPA. The selective response in OPA to Dynamic Scenes was not due to domain-general motion sensitivity or to low-level information inherited from early visual regions. Taken together, these findings suggest the novel hypothesis that OPA may support visually-guided navigation, insofar as first-person perspective motion information is useful for such navigation, while RSC and PPA support other aspects of navigation and scene recognition.