Michael O. Duff

Dopamine, uncertainty and TD learning

Substantial evidence suggests that the phasic activities of dopaminergic neurons in the primate midbrain represent a temporal difference (TD) error in predictions of future reward, with increases above and decreases below baseline consequent on positive and negative prediction errors, respectively. However, dopamine cells have very low baseline activity, which implies that the representation of these two sorts of error is asymmetric. We explore the implications of this seemingly innocuous asymmetry for the interpretation of dopaminergic firing patterns in experiments with probabilistic rewards which bring about persistent prediction errors. In particular, we show that when averaging the non-stationary prediction errors across trials, a ramping in the activity of the dopamine neurons should be apparent, whose magnitude is dependent on the learning rate. This exact phenomenon was observed in a recent experiment, though being interpreted there in antipodal terms as a within-trial encoding of uncertainty.

Procedure for electron and ion lens optimization

This letter describes a computational technique for optimal control problems arising in the synthesis of electron and ion lenses. The method provides an effective search algorithm for electrostatic and/or magnetic imaging fields with minimum aberrations. An optimized electrostatic field distribution is given as an example of application.

Assessing cultured colonies automatically

The number of colonies formed by macrophage colony-forming cells and high proliferation potential colony-forming cells was assessed by an image processor. The processor counted and sized colonies accurately, reproducibly, rapidly (2 s/dish) and objectively. The processor also measured the amount of light (in grey levels) the colonies transmitted. The optical density of a colony (the sum of its grey levels) was related to its cellularity. Thus the image processor compared both the number of colonies in samples and their cellularity. Samples of marrow containing high proliferation potential colony-forming cells of different proliferative capacity were prepared by injecting fluorouracil into mice and collecting their marrow 2-10 days later (marrow samples called FU2-FU10). These samples were cultured with one of three sources of synergistic factor titrated over seven dilutions. Colonies contained approx. 5 X 10(4) cells after 11 days culture but the way that FU2-FU10 marrow grew depended on the interval between treating donors with fluorouracil and collecting their marrow. Samples collected 2-4 days after fluorouracil formed more colonies containing more cells with small increases of synergistic factor whereas samples collected after 8-10 days did neither. It was important to culture samples of marrow with the appropriate synergistic factor for the interval after fluorouracil. Factor(s) derived from the 5637 cell line acted optimally on high proliferation potential colony-forming cells in samples collected 2-8 days after fluorouracil, and factor(s) derived from Wehi 3B cells on high proliferation potential colony-forming cells in samples collected 6-10 days after fluorouracil. Factor(s) derived from placental conditioned medium acted well on samples collected between 2 and 10 days. The proliferative capacity of samples of marrow could also be compared by estimating growth curves for high proliferation potential colony-forming cells in samples collected at successive intervals after fluorouracil.