Correlation between behavioral markers of memory retrieval and long-term potentiation (LTP) observed at regions with groups of synapses is one of the examples of an inter-connecting bridge between the fields of psychology and neuroscience. However, several mismatches between the findings at the time of learning and LTP induction have raised concerns. This provides an opportunity to re-examine the situation and arrive at the correct cellular mechanism of learning from which retrieval of memory can occur, which will enable an explanation of both the correlations and the mismatches. Since there are no cellular-level changes observed during memory retrieval, a passive reactivation of a stably maintained learning-induced mechanism is responsible for both memory and the concurrently occurring behavioral motor actions. In this context, when memories are viewed as first-person internal sensations, the derived cellular-level mechanism provides explanations for almost all the features of LTP, its correlations and mismatches.