Many aspects of higher brain function rely on two closely related capacities, inference and learning. Inference is the process of drawing conclusions from uncertain data, like forming a percept from noisy sensory information or predicting the most rewarding future outcome from the recent history of outcomes. These inferences often inform decisions that determine behavior. Learning uses experience to shape how these kinds of inference and decision processes function, often optimizing them to meet particular goals. Recent work has begun to identify how and where in the brain inference processes are implemented, particularly in the service of perceptual and reward-based decision-making. Research in my laboratory focuses on how these processes are shaped by learning to provide the flexibility a decision-maker needs to survive in a complex and dynamic world.
We use several complementary approaches to study this complex issue.
1) Quantitative measures of behavior (“psychophysics”) combined with non-invasive measures of physiological variables like pupil diameter in human subjects. These studies allow us to prototype new behavioral tasks, identify and quantify interesting behaviors, and begin to make inferences about and understand the underlying neural mechanisms.
2) Psychophysics and electrophysiology in non-human primates. These studies allow us to test directly ideas about the relationship between neural activity in a particular brain region or regions and behavior.
3) Computational modeling. These studies help to define optimal limits on behavior, characterize relationships between behavioral and neural data, and identify particular computations that can drive complex behaviors.
The goal of our work is to provide new insights into the neural mechanisms that govern complex, learned behaviors and ultimately translate these insights into new approaches to understand, diagnose, and treat disorders of learning and cognition.
Nassar MR, Bruckner R, Gold JI, Li S-C, Heekeren HR, Eppinger B: Age differences in learning emerge from an insufficient representation of uncertainty in older adults. Nature Communications 2016.
Joshi S, Li Y, Kalwani R, Gold JI: Relationships between pupil diameter and neuronal activity in the locus coeruleus, colliculi, and cingulate cortex. Neuron 89(1): 221-34, 2016.
Tsunada J, Liu ASK, Gold JI*, Cohen YE* (*contributed equally): Causal contribution of primate auditory cortex to auditory perceptual decision-making. Nature Neuroscience 19(1): 135-42, Dec 2015.
Glaze CM, Kable JW, Gold JI: Normative evidence accumulation in unpredictable environments. eLife 4, August 2015.
Liu ASK, Tsunada J, Gold JI, Cohen YE: Temporal Integration of Auditory Information Is
Invariant to Temporal Grouping Cues. eNeuro 2: 1-15, March/April 2015.
Kalwani RM, Joshi S, Gold JI: Phasic activation of individual neurons in the locus ceruleus/subceruleus complex of monkeys reflects rewarded decisions to go but not stop. J Neurosci 34(41): 2566-14, Oct 2014.
McGuire JT, Nassar MR, Gold JI, Kable JW: Functionally Dissociable Influences on Learning Rate in a Dynamic Environment. Neuron 84, 2014.
Wilson RC, Nassar MR, Gold JI: A Mixture of Delta-Rules Approximation to Bayesian Inference in Change-Point Problems. PLoS Computational Biology 9(7): 18 pp, July 2013.
Fitzgerald JK, Freedman DJ, Fanini A, Bennur S, Gold JI, Assad JA : Biased associative representations in parietal cortex. Neuron 77: 180-191, 2013.
Ding L, Gold JI: Neural correlates of perceptual decision making before, during, and after decision commitment in monkey frontal eye field. Cerebral Cortex 22(5): 1052-67, May 2012.
back to top
Last updated: 09/20/2016
The Trustees of the University of Pennsylvania