BRIEF HISTORY
For an excellent review of earlier cerebellar research, see Glickenstein 2009.
Purkinje cells were discovered just a few years before Schwann proposed his Cell Doctrine. Although it is tempting to call Purkinje’s description of the cells which bear his name the first piece of evidence in support of the Neuron Doctrine, this would be stretching the truth. In reality, anatomists at the time were still debating whether neurons in the brain stood as exceptions to the Cell Doctrine. With their long, elaborate axons and dendrites, some anatomists proposed that neurons formed a continuous, interconnectedinter-connected reticulum. Other anatomists thought that the neurofibrils were not continuous with each other but merely contacted …show more content…
In essence, he proposed that the cerebellum is a principal agent of learning motor skills. He thought that the cerebrum recruited the cerebellum in order to automate its motor commands so that a relatively simple or incomplete command from the cerebrum would be elaborated and completed by the cerebellum after motor learning. Further, he proposed that the cerebellum is a site of learning because it contains many neurons and synapses, most of which are probably modifiable, and that absent plasticity, one wouldn’t need so many neurons to merely elaborate forebrain motor commands in some routine, unchanging way. This notion of treating the immense number of neurons in the cerebellum—in particular, the granule cells—as a reservoir of modifiable elements continues to permeate theories of cerebellar cortex to this day.
Following this lecture, Giles began working out the computational features of modifiable synapses in order to explain the basic properties of classical and operant conditioning. Between the work of behaviorists like Pavlov to establish the features of associative learning, Giles’ theories of neural learning rules, and Eccles’ recently-published compendium of cerebellar anatomy and neurophysiology, the stage was set for young David Marr to tackle the problem of how exactly the cerebellum leverages its huge reservoir of granule cells to learn …show more content…
Marr’s main prediction was that parallel fiber-to-Purkinje cell synapses were strengthened during learning; in contrast, Albus postulated that the parallel fiber-to-Purkinje cell synapses would weaken (he also thought they could strengthen and that the parallel fiber-to-stellate cell synapses were also modifiable.) Albus defended his synaptic weakening argument by pointing out that the parallel fibers should be trained to emulate the post-complex spike pause and not the excitatory phase of the climbing fiber-induced complex spike, as Marr suggested. In order to emulate the post-complex spike pause, the parallel fiber synapses should weaken, and so the crucial mechanism for Albus’ theory was a weakening of the parallel fiber synapse. Thus, although they agreed that the parallel fiber-to-Purkinje cell synapse was modifiable, they disagreed over the direction of change most relevant for storing motor memories in cerebellar
For an excellent review of earlier cerebellar research, see Glickenstein 2009.
Purkinje cells were discovered just a few years before Schwann proposed his Cell Doctrine. Although it is tempting to call Purkinje’s description of the cells which bear his name the first piece of evidence in support of the Neuron Doctrine, this would be stretching the truth. In reality, anatomists at the time were still debating whether neurons in the brain stood as exceptions to the Cell Doctrine. With their long, elaborate axons and dendrites, some anatomists proposed that neurons formed a continuous, interconnectedinter-connected reticulum. Other anatomists thought that the neurofibrils were not continuous with each other but merely contacted …show more content…
In essence, he proposed that the cerebellum is a principal agent of learning motor skills. He thought that the cerebrum recruited the cerebellum in order to automate its motor commands so that a relatively simple or incomplete command from the cerebrum would be elaborated and completed by the cerebellum after motor learning. Further, he proposed that the cerebellum is a site of learning because it contains many neurons and synapses, most of which are probably modifiable, and that absent plasticity, one wouldn’t need so many neurons to merely elaborate forebrain motor commands in some routine, unchanging way. This notion of treating the immense number of neurons in the cerebellum—in particular, the granule cells—as a reservoir of modifiable elements continues to permeate theories of cerebellar cortex to this day.
Following this lecture, Giles began working out the computational features of modifiable synapses in order to explain the basic properties of classical and operant conditioning. Between the work of behaviorists like Pavlov to establish the features of associative learning, Giles’ theories of neural learning rules, and Eccles’ recently-published compendium of cerebellar anatomy and neurophysiology, the stage was set for young David Marr to tackle the problem of how exactly the cerebellum leverages its huge reservoir of granule cells to learn …show more content…
Marr’s main prediction was that parallel fiber-to-Purkinje cell synapses were strengthened during learning; in contrast, Albus postulated that the parallel fiber-to-Purkinje cell synapses would weaken (he also thought they could strengthen and that the parallel fiber-to-stellate cell synapses were also modifiable.) Albus defended his synaptic weakening argument by pointing out that the parallel fibers should be trained to emulate the post-complex spike pause and not the excitatory phase of the climbing fiber-induced complex spike, as Marr suggested. In order to emulate the post-complex spike pause, the parallel fiber synapses should weaken, and so the crucial mechanism for Albus’ theory was a weakening of the parallel fiber synapse. Thus, although they agreed that the parallel fiber-to-Purkinje cell synapse was modifiable, they disagreed over the direction of change most relevant for storing motor memories in cerebellar