Synapses are perhaps the most numerous computational elements within neural circuits. The process of chemical transmission can transform neural signals and, because synapses are plastic, these transformations can be altered over different time-scales to adjust the input-output relation of the circuit as a whole. We therefore need to assess the activity of large populations of synapses if we want to understand how neural circuits function. I will describe some experimental strategies that allow the synaptic basis of circuit function to be studied in vivo by imaging of genetically-encoded reporters (Dreosti and Lagnado, 2011). The two most promising approaches available currently are pHluorin-based reporters of synaptic vesicle fusion (Miesenböck et al., 1998; Granseth et al., 2006) and genetically-encoded calcium indicators localized to presynaptic terminals (Dreosti et al., 2009). I will illustrate how these reporters can be used to analyze circuit function with examples drawn from our work on the synaptic processing of visual information in the retina (Dreosti et al., 2011; Baden et al., 2011; Odermatt et al., 2012).