Functional Significance of Long-Term Potentiation for Sequence Learning and Prediction

Abstract

Population coding, where neurons with broad and overlapping firing rate tuning curves collectively encode information about a stimulus, is a common feature of sensory systems. We use decoding methods and measured properties of NMDA-mediated LTP induction to study the impact of long-term potentiation of synapses between the neurons of such a coding array. We find that, due to a temporal asymmetry in the induction of NMDA-mediated LTP, firing patterns in a neuronal array that initially represent the current value of a sensory input will, after training, provide an experienced-based prediction of that input instead. We compute how this prediction arises from and depends on the training experience. We also show how the encoded prediction can be used to generate learned motor sequences, such as the movement of a limb. This involves a novel form of memory recall that is driven by the motor response so that it automatically generates new information at a rate approximate for the task being performed.

When neurons form memories

“…Additional analyses revealed that the reactivation of patterns during the post-task rest period preserved some of the temporal order of neuronal activation from the task. This finding is particularly interesting in the light of theoretical work demonstrating that sequence information can be synaptically encoded and recalled by physiologically realistic learning rules [13]. Thus, the sequential reactivation during the rest period suggests that not only a given event, but also the particular sequence of events, is being consolidated.…”

Visual and Somatosensory Information about Object Shape Control Manipulative Fingertip Forces

“…That is, a set of predicted afferent signals are considered to be generated by the neural controller in conjunction with the efferent signals and are compared with the actual afferent signals (cf. Baev and Shimansky, 1992;Merfeld et al, 1993;Miall et al, 1993;Prochazka, 1993;Abbott and Blum, 1996).…”

Coding and learning of behavioral sequences

“…For the development of asymmetric place fields (Figure 1b), rate coding is sufficient [10,11,29,30] and a precise temporal code is not required. The development of asymmetric place fields [24,30] or, similarly, of visual receptive fields [31,32] could therefore be interpreted as a system-level signature of asymmetric Hebbian plasticity.…”

“…In this case, each item corresponds to one behavioral event and is represented by the firing rate of a set of active neurons. The transitions from one item to the next could then be triggered by adaptation of the active neurons [8], by synaptic depression of synapses between active neurons [9] or by proprioceptive feedback during behavior [10]. Learning is then possible by any form of asymmetric Hebbian plasticity [10,11].…”

“…The transitions from one item to the next could then be triggered by adaptation of the active neurons [8], by synaptic depression of synapses between active neurons [9] or by proprioceptive feedback during behavior [10]. Learning is then possible by any form of asymmetric Hebbian plasticity [10,11].Another theoretical possibility would be to encode behavioral sequences in neurons that are part of a slow loop with an oscillation frequency of, say, 10 Hz. In mammals, learning and retrieval of behavioral sequences probably involves the cerebellum.…”