also change the efficacy of synaptic connections between the synchroni?ed cells (9). Riehle et al. detected the increased synchrony by comparing the number of observed coincident spikes (red bars) with the number expected by chance, based o n a running estimate of rates over a longer interval (blue bar). This running comparisoll can identify exceptional synchronizations whenever they occur. Interestingly, these occurred at times when the monkey expected a possible cue, as well as when h e made a motor response. Thus, synchronization and rate coding can occur separately or together (8, 9 , 12, 14, 15).Several other approaches can detect coherent activity in neural populations at mil-lisecol~d resolution. Standard cross-correlations between simultaneously recorded neurons commonly reveal synchronous activity, and time-dependent cross-correlation measures (1 l ) have revealed that synchronous firing can be rapidly modulated with hehavior, even without changes in firing rates (1 2).Other studies have found evidence for precisely timed patterns occurring among neural groups in a bella\~ioral situation (5). Simulations have slloaln that such synchrony can he preserved in chains of neurons with realistic synaptic connections ( 13). Another for111 of synchronked activity in neural popillations is the widespread periodic oscillations seen in visual cortex neurons, which has been suggested to subserve a binding function (9), a suggestion potentially applicable to coordination of motor responses (14, 15). Another approach uses the "gravity" method to identify groups of neurons that tend to fire in synchrony: If n neurons are located in ndinlensiollal space, and their spikes are endowed with a transient "charge," those cells that fire synchronously tend to be attracted and form identifiable clusters (6).Although all of these algorithms can detect the existence of precise tenlporal structure in neural activity, this does not yet establish their function as a temporal code. W h a t is needed first is some demonstration that synchronization occurs reliably under particular behavioral conditions. T h e accumulating evidence is suggesti.~re (5-9, 12, 14) hut still leaves the exact function unproven. Establishing the fi~nctional mechanism inay not he a matter of finding tighter correlations with behavioral events; for example, holographic lnechanislns code distributed inforlnation in terms of phase relations rather than literal representations (2). Skeptics can still argue that tlle telnporal events revealed by these methods are epiphenomena or products of tlle statistical models, and that anything telnporal coding can do, population rate coding can do as well (16). Moreover, there are open questions abililt how temporal codes are established and how they interact with rate coding (8, 15). These issues can now be investigated with the tools at hand: Multiunit recordil~gs can be analyed with these algorithms, and the detected events can be related to behavior. Neural netalork sitnulations call also help to demonstrate how the pu...