The relevance of backpropagating action potentials to synaptic integration is no longer in question. When a spike travels back into the dendritic arbour, it can affect the functional properties of active synapses in several ways. For example, backpropagating action potentials can lead to potentiation or depression of synaptic transmission depending on the exact timing of spikes and synaptic potentials. As the interaction between ongoing activity and backpropagating action potentials has profound implications for the integration of synaptic inputs, it is not surprising that the study of this phenomenon has gained significant notoriety. Three papers published over the past few weeks attest to the rapid development of this area.

Stuart and Häusser investigated the influence of excitatory postsynaptic potentials (EPSPs) on the amplitude of backpropagating spikes in cortical neurons. They found that an EPSP at distal dendritic sites can increase spike amplitude by three- to fourfold, but for this increase to occur, both EPSPs and backpropagating spikes must occur within a small time window. In other words, the distal dendrites can detect the coincidence of backpropagating spikes and EPSPs, a property that is likely to be involved in the induction of plastic changes at the synapse.

Whereas EPSPs can influence backpropagating spike amplitude in the distal dendrites, Häusser et al. showed that the large conductances activated during the spike at the soma reduce coincident EPSPs in a cell- and synapse-specific manner. This shunting effect of action potentials depends on both the duration and location of the synaptic input, providing a mechanism for distal and NMDA-receptor mediated inputs to contribute disproportionately to synaptic integration during spike firing.

Although the data obtained in brain slices are fairly compelling, it is important to know whether backpropagating action potentials also affect synaptic function in vivo. This is a difficult question to answer, but Quirk et al. have taken an important first step. They recorded extracellular spikes in the hippocampi of freely moving rats and observed that the amplitude of the spikes decreased with ongoing behaviour, a reduction that was countered by the rat's experience with the experimental environment. The evidence that the decrease in extracellular spike amplitude is related to the attenuation of backpropagating action potentials remains indirect. However, this study is an initial attempt to link the abundant in vitro observations and the role of backpropagating spikes in the behaving animal.