 In this video, each pixel is connected as an input to a phase change synapse, here 10,000 synapses per neuron. The Watson avatar and IBM research patterns appear at certain time instances, whereas noise appears in the rest of the pixels. The synapses of the corresponding or level-tuned neurons learn the patterns without any further computation or in an unsupervised way by increasing the conductance of the nanodevices that correspond to the Watson and IBM research images and decreasing the conductance of the nanodevices that correspond to the noise. Simply put, without any human intervention, we have mimicked the functionality of neurons and synapses to distinguish the Watson and IBM research patterns and detect their arrival times on their own, which is a significant step forward in the development of energy-efficient, ultra-dense integrated neuromorphic technologies for applications in cognitive computing.