Equal Numbers Of Neurons Research Articles

Determining the Number of Neurons in Artificial Neural Networks for Approximation, Trained with Algorithms Using the Jacobi Matrix

How can we determine the optimal number of neurons when constructing an artificial neural network? This is one of the most frequently asked questions when working with this type of artificial intelligence. Experience has brought the understanding that it takes an individual approach for each task to specify the number of neurons. Our method is based on the requirement of algorithms looking for a minimum of functions of type 𝑺􁈺𝒛􁈻 􀵌 Σ 􁈾𝝋𝒊 𝒎 􁈺𝒛 􁈻􁈿𝟐 𝒊􀭀𝟏 that satisfy the inequality 𝒑 􀵑 𝒎, where p is the dimensionality of the argument z, and m is the number of functions. Formulas for an upper limit of the required neurons are proposed for networks with one hidden layer and for networks with r hidden layers with an equal number of neurons.

Social Signals in Primate Orbitofrontal Cortex

Primate evolution produced an increased capacity to respond flexibly to varying social contexts as well as expansion of the prefrontal cortex. Despite this association, how prefrontal neurons respond to social information remains virtually unknown. People with damage to their orbitofrontal cortex (OFC) struggle to recognize facial expressions, make poor social judgments, and frequently make social faux pas. Here we test explicitly whether neurons in primate OFC signal social information and, if so, how such signals compare with responses to primary fluid rewards. We find that OFC neurons distinguish images that belong to socially defined categories, such as female perinea and faces, as well as the social dominance of those faces. These modulations signaled both how much monkeys valued these pictures and their interest in continuing to view them. Far more neurons signaled social category than signaled fluid value, despite the stronger impact of fluid reward on monkeys' choices. These findings indicate that OFC represents both the motivational value and attentional priority of other individuals, thus contributing to both the acquisition of information about others and subsequent social decisions. Our results betray a fundamental disconnect between preferences expressed through overt choice, which were primarily driven by the desire for more fluid, and preferential neuronal processing, which favored social computations.

TRPA1 contributes to specific mechanically activated currents and sensory neuron mechanical hypersensitivity

The mechanosensory role of TRPA1 and its contribution to mechanical hypersensitivity in sensory neurons remains enigmatic. We elucidated this role by recording mechanically activated currents in conjunction with TRPA1 over- and under-expression and selective pharmacology. First, we established that TRPA1 transcript, protein and functional expression are more abundant in smaller-diameter neurons than larger-diameter neurons, allowing comparison of two different neuronal populations. Utilising whole cell patch clamping, we applied calibrated displacements to neurites of dorsal root ganglion (DRG) neurons in short-term culture and recorded mechanically activated currents termed intermediately (IAMCs), rapidly (RAMCs) or slowly adapting (SAMCs). Trpa1 deletion (–/–) significantly reduced maximum IAMC amplitude by 43% in small-diameter neurons compared with wild-type (+/+) neurons. All other mechanically activated currents in small- and large-diameter Trpa1−/− neurons were unaltered. Seventy-three per cent of Trpa1+/+ small-diameter neurons responding to the TRPA1 agonist allyl-isothiocyanate (AITC) displayed IAMCs to neurite displacement, which were significantly enhanced after AITC addition. The TRPA1 antagonist HC-030031 significantly decreased Trpa1+/+ IAMC amplitudes, but only in AITC responsive neurons. Using a transfection system we also showed TRPA1 over-expression in Trpa1+/+ small-diameter neurons increases IAMC amplitude, an effect reversed by HC-030031. Furthermore, TRPA1 introduction into Trpa1−/− small-diameter neurons restored IAMC amplitudes to Trpa1+/+ levels, which was subsequently reversed by HC-030031. In summary our data demonstrate TRPA1 makes a contribution to normal mechanosensation in a specific subset of DRG neurons. Furthermore, they also provide new evidence illustrating mechanisms by which sensitisation or over-expression of TRPA1 enhances nociceptor mechanosensitivity. Overall, these findings suggest TRPA1 has the capacity to tune neuronal mechanosensitivity depending on its degree of activation or expression.

Quantitative and neurogenic analysis of the total population and subpopulations of neurons defined by axon projection in the superficial dorsal horn of the rat lumbar spinal cord.

The total neuron population of the superficial dorsal horn (SDH), i.e., laminae I and II, was quantitated in Nissl preparations of spinal segment L1 in the rat. Subpopulations of the SDH, defined by axon projection, were quantitated following strategic intraspinal injections of dual retrograde tracers (Fluoro-Gold and true blue). These methods were used in conjunction with [3H]thymidine (delivered in utero) autoradiography for neurogenic pattern analysis. Following stereological correction, each dorsal horn in spinal segment L1 contained 11 neurons in lamina I and 42.6 neurons in lamina II per 10-microm transverse section. Neurons with long projections, i.e., neurons with projections rostral to spinal segment T5, were only slightly more numerous in lamina I than in lamina II. These neurons made up 34% of the total neuron population in lamina I and 7.0% in lamina II. Most of these neurons did not demonstrate descending connections, and many (presumed supraspinal projection neurons) did not demonstrate short, ascending, intersegmental connections. Neurons with short propriospinal projections, i.e., neurons with connections caudal to spinal segment T5, made up approximately half of the total neuron population in both lamina I and lamina II: 55% and 52%, respectively. Of these, 79% had both short ascending and descending projections; the remaining 21% had only descending projections. Neurons that were not labeled with retrograde tracers (presumed local circuit cells) represented 11% of the neurons in lamina I and 41% in lamina II. Neurogenesis in the SDH proceeded along an axon-length gradient, whereby neurons with the longest axons completed neurogenesis first, and those with the shortest completed neurogenesis last. The generation of both propriospinal and supraspinal projection neurons began on embryonic day 13 (E13). Nearly equal numbers of neurons in this group were generated in laminae I and II through E14. On E15, neuron production slowed in lamina I and accelerated in lamina II as local circuit neurons and the remaining propriospinal neurons were generated. Neuron production ceased simultaneously in both lamina I and lamina II on E16.

Dopamine enhancement and depression of glutamate-regulated calcium and electrical activity in hypothalamic neurons.

1. The neurotransmitter dopamine is found throughout the hypothalamus both in cell bodies and in axons originating from intra- and extrahypothalamic sources. To study the mechanisms of action of dopamine on cultured rat hypothalamic neurons, particularly in relation to Ca2+ regulation, we used Ca2+ digital imaging with fura-2 and whole cell patch-clamp recording. We focused on the modulatory actions of dopamine on glutamate. 2. Dopamine administration had little or no independent effect on intracellular Ca2+. However, in the presence of tetrodotoxin to block action potentials and action-potential-dependent transmitter release, dopamine (10 microM for 2-3 min) caused an increase in glutamate-evoked Ca2+ rises in 22% of 64 neurons and depressed glutamate-evoked Ca2+ rises in an equal number of neurons. Shorter exposure to dopamine reduced the number of responding cells. 3. Dopamine application to neurons with an elevated Ca2+ due to synaptic release of glutamate (in the absence of tetrodotoxin) generally caused a decrease in Ca2+ levels (40% of 106 neurons), but sometimes increased cytosolic Ca2+ (10% of 106 neurons). That dopamine influenced cells differently in conditions of spontaneous activity compared with evoked activity may be due to dopamine effects on presynaptic receptors detected under conditions of ongoing synaptic release of glutamate. 4. Dopamine modulation of glutamate responses was detected at early stages of neuronal development (embryonic day 18 after 2 days in vitro) and also after 60 days in vitro. 5. The D1, D2, and D3 dopamine receptor agonists SKF38393, quinpirole, and 7-OH-DPAT (+/- 7 hydroxy-dipropylaminotetralin) caused a reduction in Ca2+ levels raised by endogenous glutamate release or evoked by exogenous glutamate application. 6. To block the actions of dopamine released by hypothalamic neurons, D1 and D2 dopamine receptor antagonists were used. As with dopamine, dopamine antagonists had no effect on intracellular Ca2+ during glutamate receptor blockade. In the absence of glutamate receptor block, the D1 antagonist SCH23390 (1 microM) reduced Ca2+ in responding cells; in contrast, the D2 antagonist eticlopride (1 microM) generated a delayed increase in Ca2+ levels. 7. Dopamine is known to activate second messengers through G proteins independent of changes in membrane potential or input resistance. Whole cell recording was used to demonstrate that, parallel to the modulation of Ca2+, dopamine exerted a dramatic change in glutamate-mediated electrical activity, generally depressing activity and hyperpolarizing the membrane potential (8 of 15 neurons). In a smaller number of neurons (5 of 15), dopamine enhanced glutamate-mediated excitatory activity. 8. Dopamine-evoked changes in membrane potential were in part mediated through modulation of glutamate actions. Dopamine depressed glutamate-evoked currents in a dose-dependent fashion, with Hill slopes in individual neurons ranging from 0.3 to 0.6. Dopamine could also evoke a direct hyperpolarizing action on hypothalamic neurons in the presence of tetrodotoxin or glutamate receptor blockers, at least in part by opening K+ channels. 9. Glutamate plays an important role as a primary excitatory transmitter within the hypothalamus. Our data support the hypothesis that a major mechanism of dopamine's influence on hypothalamic neurons involves the modulation of glutamate's excitatory action, mostly by inhibition. This is consistent with the hypothesis that modulation of glutamate activity may be an important mechanism of dopamine action throughout the nervous system.

Gender differences in neurotransmitter expression in the rat superior cervical ganglion

Sexual dimorphism of neuron number has been observed in several areas in the central and peripheral nervous system. In many of these areas enhanced neuron survival exists in males during the period of naturally occurring cell death. This has been attributed to high levels of circulating testosterone in the perinatal period. The superior cervical ganglion (SCG) of the rat exhibits this sexual dimorphism. The difference in neuron numbers is established by two weeks postnatally and precedes the differences in body weight and sympathetic target mass between the sexes. At this two week time point, fewer SCG neurons in the female rat must supply neurotransmitter to the same mass of sympathetic target as in the male. The present study examined some of the mechanisms used by neurons in the SCGs of male and female rats to compensate in supplying neurotransmitter to their targets. At birth, the SCGs of male and female rats contain equal numbers of neurons. There is also no sex difference at this time in the content of norepinephrine (NE) in these neurons or in the enzyme activity of tyrosine hydroxylase (TH). However, a sex difference does exist in the expression of TH-mRNA, with SCG neurons in female expressing more TH-mRNA than males. At this time, there is no sex difference in either the total body weight of males and females or in the mass of sympathetic target organs. During the first two postnatal weeks, natural neuron death in the SCG results in the loss of significantly more neurons in females than in males. At the end of the period of cell death, neurons in females continue to express more TH-mRNA, and at this time both TH enzyme activity and NE content per neuron are also higher in females. Since the adult sex difference in body weight and sympathetic target mass has not yet been established, the same amount of target mass is innervated by fewer neurons in females. In the adult, the sex difference in SCG neuron number is maintained. However, both overall body weight and sympathetic target mass are significantly greater in males. At this time expression of TH-mRNA is greater in SCG neurons of males, while both TH enzyme activity and NE content per neuron are equal in males and females. One of the challenges presented to the developing nervous system is to match a population of neurons with its targets. These results demonstrate that there are changes in both cell number and neurotransmitter gene expression in the developing SCG of both males and females before there is any sex difference in the amount of target to be innervated. We conclude that there is some mechanism, mediated by gonadal steroids, which anticipates the demand of the target and regulates the establishment of cell number and the development of neurotransmitter metabolism.

Responses in the diagonal band of Broca, adjacent septal nuclei and the islands of Calleja of cats to stimulation of the subcallosal fornix, medial basal hypothalamus and medial forebrain bundle

The projection of neurons in the septal nuclei and the insula magna of the islands of Callaja (IC) was explored together with their response to stimulation of the fornix. The septal nuclei all contained neurons projecting in the medial forebrain bundle (MFB). Only the diagonal band of Broca (DBB) and the lateral septal nucleus (LS) contained many neurons projecting toward the medial basal hypothalamus (MBH). The spatial distribution of neurons excited by stimulation of the fornix in the DBB was almost identical with the distribution of neurons projecting toward the MBH and there was considerable overlap (10/28 cells). In the medial septal nucleus the spatial distribution of neurons excited by stimulation of the fornix and neurons projecting in the MFB was similar and there was considerable overlap (6/21 cells). The connectivity of the IC resembled that of the MS but there was little overlap between the neurons excited by fornix stimulation and those projecting in the MFB (1/27 cells). In the LS there were almost equal numbers of neurons projecting in the MFB and toward the MBH but there was very little input from the fornix. Neurons were significantly more often excited by stimulation of the lateral fornix, carrying axons from the subiculum, than they were by medial stimulation exciting axons from Ammon's horn. Axons projecting toward the MBH or in the MFB had conduction velocities less than 1 m/s.

Size and density of glial and neuronal cells within the cerebral neocortex of various insectivorian species

Morphometric measurements were done on frontal sections through the somatosensory neocortex of various insectivorian species. All measured parameters varied with the size of animals; there was a better correlation with the ventriculartopial brain wall thickness than with the brain weight. The following rules were evaluated: with increasing brain wall thickness, 1) lamina I becomes thinner; 2) the nuclei of both neuronal and glial cells become larger; 3) the volume density of neuronal cells decreases greatly; 4) the volume density of glial cells increases slightly; and 5) as a result, the glia:neuron index increases markedly. There was no equal number of neurons under a unit surface area in the cortices of any species studied. Developmental processes that might account for the above-mentioned rules are discussed in this report.

Substance P-induced depolarization in sympathetic neurons: not simple K-inactivation

Ionic mechanisms underlying substance P-induced depolarization of the inferior mesenteric ganglion cells of the guinea pig were analyzed by means of microelectrode methods. When the membrane potential was manually clamped at the resting level, substance P caused, in about equal number of neurons, increases and decreases of neuronal input resistance. In the majority of the cells tested the amplitude of substance P-induced depolarization was increased when the membrane was hyperpolarized to the level of E k; it was markedly reduced in a Na +-free media. These results suggest that substance P causes depolarization by simultaneously increasing Na + and decreasing K + permeability.