next
Cold Spring Harbor Laboratory
Please visit the Brody Lab home page


Click to Enlarge Graphic


Firing rates of 6 different neurons during a somatosensory working memory task. A vibration stimulus is applied to a fingertip for 500 ms (grey box in each panel is time of stimulation). Neurons in prefrontal cortex respond with spike firing rates that depend on the frequency of vibration (colors indicate different vibration stimuli, as indicated by color code at top). The two neurons in the middle row show stimulus-dependent firing throughout the 3 seconds that the animal was trained to remember the frequency for. If we listen to these neurons, we can tell what it is that the animal is remembering.
Carlos D. Brody
Adjunct Associate Professor
Ph.D., California Institute of Technology, 1998
Computational neuroscience; psychophysics of short-term memory; neural data analysis; computation with spiking neural networks

email brody@cshl.edu, phone (516) 367-6902 , fax (516) 367-8389

Carlos BrodyYou look up a number in the phone book, you remember it for a few seconds until you dial it, and then you let it go, forgetting it. This is a classic example of working memory, the ability to keep something in mind for a few seconds. Working memory is thought to be fundamental to many cognitive processes, including, for example, holding a conversation, playing music, or even thinking rationally. What are the neural mechanisms behind it?

In collaboration with experimental neurophysiologists, we have studied responses of prefrontal cortical neurons in monkeys trained to perform a working memory task. Examples of the responses of six such neurons are shown in the figure opposite. The neurons in the middle row are a clear neural correlate of working memory: their firing rate throughout memory periods reflects the remembered stimulus.

The research in my lab focuses on possible mechanisms behind such responses. One approach we are pursuing is to build computational models of networks of neurons that display maintained, stimulus-dependent responses. The models can help us pinpoint the crucial assumptions behind our current hypotheses of how memory activity can arise. The models also make predictions about our capacities to keep sensory stimuli in short-term memory. A second approach we are pursuing is to test these neural network model-based predictions by carrying out psychophysical experiments with humans. Our goal is to bridge the gap between computational neural network-level models and psychophysically tested human behavior.


Please visit the Brody Lab home page.

Selected Publications

Machens, C.K., R. Romo, and C.D. Brody. 2005. Flexible control of mutual inhibition: a neural model of two-interval discrimination. Science 307: 1121–1124.

Hopfield, J.J., and C.D. Brody. 2004. Learning rules and network repair in spike-timing-based computation networks. Proc. Natl. Acad. Sci. USA 101: 337–342.

Brody, C.D., and J.J. Hopfield. 2003. Simple networks for spike-timing-based computation, with application to olfactory processing. Neuron 37: 843–852.

Hopfield, J.J., and C.D. Brody. 2001. What is a moment? Transient synchrony as a collective mechanism for spatiotemporal integration. Proc. Natl. Acad. Sci. USA 98: 1282–1287.

Romo, R., C.D. Brody, A. Hernandez, and L. Lemus. 1999. Neuronal correlates of parametric working memory in the prefrontal cortex. Nature 399: 470–473.