Economic choice is specifically disrupted in mental and neurological disorders such as major depression,
frontotemporal dementia, and drug addiction. Established literature links this behavior to the orbitofrontal cortex
(OFC). In recent years, my lab has examined neuronal activity in the OFC of monkeys and mice choosing
between different juice types. Our studies revealed the presence of different cell groups: offer value cells
encoding the value of individual offers; choice outcome cells encoding the identity of the chosen offer or the
chosen action; and chosen value cells. In a computational sense, these variables capture both the input (offer
value) and the output (choice outcome, chosen value) of the choice process, suggesting that the cell groups
identified in OFC constitute the building blocks of a decision circuit. Indirect evidence supports this hypothesis.
However, the anatomical organization of this circuit and the contributions of different cell groups are not known.
Consequently, the decision mechanisms remain poorly understood. To shed light on these fundamental
questions, we developed a preparation in which mice perform economic choices while we record from OFC
using two-photon (2P) calcium (Ca2+) imaging through a GRIN lens. Longitudinal recordings showed that the
encoding of specific decision variables by individual cells is very stable. Moreover, the representation of
decision variables differs significantly across cortical layers: input variables populate mostly layers 2/3 (L2/3),
while spatial and output variables populate mostly layer 5 (L5). These results support the notion of a stable
decision circuit and provide a glimpse of its anatomical organization. However, to truly unravel the decision
mechanisms, one needs to have a detailed understanding of how different cell groups connect with each other,
how they influence each other’s activity, and how the activity of each cell group affects choices. Here we
propose to address these questions using a combination of 2P imaging and single cell optogenetics. In Aim 1,
we will assess the effective connectivity between cell groups. Mice will (a) be implanted with a GRIN lens
accessing L2/3 and/or L5, (b) express a green Ca2+ indicator and a red shifted channelrhodopsin (ChR), and
(c) perform choices under the microscope. In each mouse, we will image ~500 cells. First, we will assess the
variable encoded by each cell. In separate sessions, while animals rest, we will holographically stimulate one
specific cell group while imaging the entire population. Measuring the activity evoked by the stimulation, we will
assess the effective connectivity between different cell groups. In Aim 2, we will use the same preparation, but
we will deliver the stimulation while mice perform choices (in half of the trials). We will measure how optical
stimulation affects performance, and thus assess the causal links between individual cell groups and choices.
Relevance to R21. These experiments are innovative and technically challenging (high risk), but they can
potentially provide unprecedented insights into the mechanisms underlying economic decisions (high reward).
The proposal is exploratory, but we have the background and expertise to successfully conduct this research.