In humans, there are just 10^3 skeletal muscles in the body (this is a round up from 750 skeletal muscles for the sake of simplicity). This means that every muscle (involved in volitional control) would have 30,000 Purkinje neurons available given that the total number of Purkinje cells per cerebellum is 30 x 10^6 (Huang 2008), and each Purkinje neuron on average is supplied by 2,000 granular neurons (Herculano-Houzel 2009), which carry signals from the neocortex, the brain stem, and the spinal cord via the mossy fibres [note the percentage of cerebellum dedicated to controlling the smooth and cardiac muscles as well as the glands awaits verification]. Using a firing rate-code for the execution of both oculo- and skeletomotor movements (Kheradmand and Zee 2011; Miles et al. 1980ab; Miles and Lisberger 1981; Noda and Fujikado 1987ab; Sauerbrei, Siapas et al. 2015; Tehovnik et al. 2021), which is the code used by skeletomotor and ocular efferents, the efference-copy representation stored within Purkinje neurons (Bell et al. 1997; De Zeeuw 2021; Giovannucci et al. 2017; Loyola et al. 2019; Shadmehr 2020; Tehovnik et al. 2021; Wang et al. 2023) is set through learning via connections from the neocortex, the brain stem, and the spinal cord. This representation finalized allows for variables such as movement accuracy, latency, speed, and acceleration to be optimized (Miles et al. 1980ab; Miles and Lisberger 1981; Sauerbrei, Siapas et al. 2015). Indeed, it is the simple spikes of the Purkinje neurons, receptive to information supplied by granular cells (Heck and Sultan 2002; Sultan and Heck 2003), that must have their gains adjusted by the complex-spike input from the inferior olive that automates all behavioral responses from the vestibulo-ocular reflex to eyeblink conditioning to running down an Olympic track to giving a presidential address and to deducing mc^2 (Tehovnik, Hasanbegović, Chen 2024). All these outputs depend on extensive training, some from hours to days and others from years to decades. The ratio of Purkinje neurons to the number of skeletal muscles of 30,000 to 1 highlights how serious the human brain is about learning (Hebb 1949, 1961, 1968). A mouse has a comparable number of skeletal muscles to that of humans, but the ratio of Purkinje neurons to the number of muscles is significantly lower at 200 to 1 (200,000 Purkinje neurons/1000 muscles, Birkisdóttir et al. 2023), but to be fair mice only have to live for 3 to 4 years, whereas humans (today) live up to 100 years, which requires lots more learning that needs to be put under automatic control (Kahneman 2011).
- Badges
- Science topic