Patterns of ocean current variability are examined on the West Florida Shelf by a neural network analysis based on the self-organizing map (SOM), using time series of moored velocity data that span the interval October 1998--September 2001. Three characteristic spatial patterns are extracted in a 3 ¿ 4 SOM array: spatially coherent southeastward and northwestward flow patterns with strong currents and a transition pattern of weak currents. On the synoptic weather timescale the variations of these patterns are coherent with the local winds. On the seasonal timescale the variations of the patterns are coherent with both the local winds and complementary sea surface temperature patterns. The currents are predominantly southeastward during fall--winter months (from October to March) and northwestward during summer months (from June to September). The spatial patterns extracted by the (nonlinear) SOM method are asymmetric, a feature that is not captured by the (linear) empirical orthogonal function method. Thus we find for the synoptic weather and longer timescales that (1) southeastward currents are generally stronger than northwestward currents, (2) the coastal jet axis is located further offshore for southeastward currents than for northwestward currents, and (3) the velocity vector rotations with depth are larger in shallower water when the currents are southeastward relative to when the currents are northwestward.