Stephanie Palmer, PhD

I study how populations of neurons collectively encode information present in their inputs and how they perform computations on these signals. The brain performs several classes of computation including signal comparison, prediction, error correction, and learning. To investigate these phenomena, I work with experimentalists on a variety of systems: predictive coding in the retina and visual cortex of the rodent, motion coding in area MT, and temporal coding in the zebra finch song system. From these studies, several general principles have emerged, which guide my current research: the hypothesis that neurons are optimized to predict their future inputs, that information in neural populations is represented combinatorially, and that coding in sensori-motor systems is highly dynamic and behaviorally dependent. By working closely with experimentalists, we constrain and test these theories of neural population coding with detailed measurements.

Princeton University
Princeton, NJ
Postdoctoral - Neuroscience
2012

University of California San Francisco
San Francisco, CA
Postdoctoral - Neuroscience
2005

University of Oxford
England, UK
DPhil - Theoretical Physics
2001

Michigan State University
East Lansing, MI
BS - Chemical Physics
1997

Stimulus invariant aspects of the retinal code drive discriminability of natural scenes.
Stimulus invariant aspects of the retinal code drive discriminability of natural scenes. bioRxiv. 2023 Aug 12.
PMID: 37609259

Learning low-dimensional generalizable natural features from retina using a U-net.
Learning low-dimensional generalizable natural features from retina using a U-net. Adv Neural Inf Process Syst. 2022 Dec; 35:11355-11368.
PMID: 37362058

Gaussian Information Bottleneck and the Non-Perturbative Renormalization Group.
Gaussian Information Bottleneck and the Non-Perturbative Renormalization Group. New J Phys. 2022 Mar; 24(3).
PMID: 35368649

Spatially displaced excitation contributes to the encoding of interrupted motion by a retinal direction-selective circuit.
Spatially displaced excitation contributes to the encoding of interrupted motion by a retinal direction-selective circuit. Elife. 2021 06 07; 10.
PMID: 34096504

Maximally efficient prediction in the early fly visual system may support evasive flight maneuvers.
Maximally efficient prediction in the early fly visual system may support evasive flight maneuvers. PLoS Comput Biol. 2021 05; 17(5):e1008965.
PMID: 34014926

Optimal prediction with resource constraints using the information bottleneck.
Optimal prediction with resource constraints using the information bottleneck. PLoS Comput Biol. 2021 03; 17(3):e1008743.
PMID: 33684112

Variable but not random: temporal pattern coding in a songbird brain area necessary for song modification.
Variable but not random: temporal pattern coding in a songbird brain area necessary for song modification. J Neurophysiol. 2021 02 01; 125(2):540-555.
PMID: 33296616

Supervised learning through physical changes in a mechanical system.
Supervised learning through physical changes in a mechanical system. Proc Natl Acad Sci U S A. 2020 06 30; 117(26):14843-14850.
PMID: 32546522

Nonlinear mixed selectivity supports reliable neural computation.
Nonlinear mixed selectivity supports reliable neural computation. PLoS Comput Biol. 2020 02; 16(2):e1007544.
PMID: 32069273

Aristaless Controls Butterfly Wing Color Variation Used in Mimicry and Mate Choice.
Aristaless Controls Butterfly Wing Color Variation Used in Mimicry and Mate Choice. Curr Biol. 2018 11 05; 28(21):3469-3474.e4.
PMID: 30415702

View All Publications