Nomenclature of optimal BMI: slim's the word.
Explore the source record for details and available documents.
Body Mass Index↗
Biomedical subjects
M J Tovee
Publications and source records attributed to M J Tovee.
The representational capacity of the distributed encoding of information provided by populations of neurons in primate temporal visual cortex.
It has been shown that it is possible to read, from the firing rates of just a small population of neurons, the code that is used in the macaque temporal lobe visual cortex to distinguish between different faces being looked at. To analyse the information provided by populations of single neurons in the primate temporal cortical visual areas, the responses of a population of 14 neurons to 20 visual stimuli were analysed in a macaque performing a visual fixation task. The population of neurons analysed responded primarily to faces, and the stimuli utilised were all human and monkey faces. Each neuron had its own response profile to the different members of the stimulus set. The mean response of each neuron to each stimulus in the set was calculated from a fraction of the ten trials of data available for every stimulus. From the remaining data, it was possible to calculate, for any population response vector, the relative likelihoods that it had been elicited by each of the stimuli in the set. By comparison with the stimuli actually shown, the mean percentage correct identification was computed and also the mean information about the stimuli, in bits, that the population of neurons carried on a single trial. When the decoding algorithm used for this calculation approximated an optimal, Bayesian estimate of the relative likelihoods, the percentage correct increased from 14% correct (chance was 5% correct) with one neuron to 67% with 14 neurons. The information conveyed by the population of neurons increased approximately linearly from 0.33 bits with one neuron to 2.77 bits with 14 neurons. This leads to the important conclusion that the number of stimuli that can be encoded by a population of neurons in this part of the visual system increases approximately exponentially as the number of cells in the sample increases (in that the log of the number of stimuli increases almost linearly). This is in contrast to a local encoding scheme (of "grandmother" cells), in which the number of stimuli encoded increases linearly with the number of cells in the sample. Thus one of the potentially important properties of distributed representations, an exponential increase in the number of stimuli that can be represented, has been demonstrated in the brain with this population of neurons. When the algorithm used for estimating stimulus likelihood was as simple as could be easily implemented by neurons receiving the population's output (based on just the dot product between the population response vector and each mean response vector), it was still found that the 14-neuron population produced 66% correct guesses and conveyed 2.30 bits of information, or 83% of the information that could be extracted with the nearly optimal procedure. It was also shown that, although there was some redundancy in the representation (with each neuron contributing to the information carried by the whole population 60% of the information it carried alone, rather than 100%), this is due to the fact that the number of stimuli in the set was limited (it was 20). The data are consistent with minimal redundancy for sufficiently large and diverse sets of stimuli. The implication for brain connectivity of the distributed encoding scheme, which was demonstrated here in the case of faces, is that a neuron can receive a great deal of information about what is encoded by a large population of neurons if it is able to receive its inputs from a random subset of these neurons, even of limited numbers (e.g. hundreds).
Algorithms↗
Information in the neuronal representation of individual stimuli in the primate temporal visual cortex.
To analyze the information provided about individual visual stimuli in the responses of single neurons in the primate temporal lobe visual cortex, neuronal responses to a set of 65 visual stimuli were recorded in macaques performing a visual fixation task and analyzed using information theoretical measures. The population of neurons analyzed responded primarily to faces. The stimuli included 23 faces and 42 nonface images of real-world scenes, so that the function of this brain region could be analyzed when it was processing relatively natural scenes. It was found that for the majority of the neurons significant amounts of information were reflected about which of several of the 23 faces had been seen. Thus the representation was not local, for in a local representation almost all the information available can be obtained when the single stimulus to which the neuron responds best is shown. It is shown that the information available about any one stimulus depended on how different (for example, how many standard deviations) the response to that stimulus was from the average response to all stimuli. This was the case for responses below the average response as well as above. It is shown that the fraction of information carried by the low firing rates of a cell was large--much larger than that carried by the high firing rates. Part of the reason for this is that the probability distribution of different firing rates is biased toward low values (though with fewer very low values than would be predicted by an exponential distribution). Another factor is that the variability of the response is large at intermediate and high firing rates. Another finding is that at short sampling intervals (such as 20 ms) the neurons code information efficiently, by effectively acting as binary variables and behaving less noisily than would be expected of a Poisson process.
Action Potentials↗
Rapid visual learning in neurones of the primate temporal visual cortex.
The human visual system can learn to recognize visual stimuli rapidly. For example, humans can accurately reconstruct meaningful objects out of fragmentary evidence, once they have seen the same object in its unambiguous form. The anterior temporal cortical areas of macaques contain some neurones with invariant visual responses which appear to provide a representation of complex patterns and objects, such as faces. Remarkably, these neurones show an enhancement of response after brief (e.g. 5 s) exposure to the unambiguous stimulus, an effect that appears to reflect the neural basis of the rapid perceptual learning seen in humans.
Animals↗
The responses of single neurons in the temporal visual cortical areas of the macaque when more than one stimulus is present in the receptive field.
Neurons in the temporal visual cortical areas of primates have large receptive fields, which can show considerable selectivity for what the stimulus is irrespective of exactly where it is in the visual field. This is called translation invariance. However, such results have been found when there is only one stimulus in the visual field. The question arises of how the visual system operates in a cluttered environment. To investigate this we measured the responses of neurons with face-selective responses in the cortex in the anterior part of the superior temporal sulcus of rhesus macaques performing a visual fixation task. We found that the response of neurons to an effective face centred 8.5 degrees from the fovea was decreased to 71% if an ineffective face stimulus for that cell was present at the fovea. In a similar way, introduction of a parafoveal ineffective face stimulus decreased the responses of these neurons to an effective face stimulus at the fovea to 75%. In addition to these interactions, it was found that an effective stimulus object at the fovea produced a larger response than when it was parafoveal, and that this weighting towards an object at the fovea was also seen when more than one object was present in the visual field. The implication of this weighting of the responses of neurons towards objects at the fovea, even in an environment with more than one object present, is that the output of the visual system provides information to subsequent systems particularly about objects at the fovea, so that learning about these objects (and less about other objects elsewhere in the visual field) is facilitated.
Animals↗
Sparseness of the neuronal representation of stimuli in the primate temporal visual cortex.
1. To analyze the selectivity and the sparseness of firing to visual stimuli of single neurons in the primate temporal cortical visual area, neuronal responses were measured to a set of 68 visual stimuli in macaques performing a visual fixation task. The population of neurons analyzed had responses that occurred primarily to faces. The stimuli included 23 faces, and 45 nonface images of real-world scenes, so that the function of this brain region could be analyzed when it was processing natural scenes. 2. The neurons were selected to meet the previously used criteria of face selectivity by responding more than twice as much to the optimal face as to the optimal nonface stimulus in the set. Application of information theoretic analyses to the responses of these neurons confirmed that their responses contained much more information about which of 20 face stimuli had been seen (on average 0.4 bits) than about which (of 20) nonface stimuli had been seen (on average 0.07 bits). 3. The sparseness of the representation of a scene or object provided by each of these neurons (which can be thought of as the proportion of stimuli to which the neuron responds, and which is fundamental to understanding the network operation of the system) can be defined as [formula: see text] where ri is the firing rate to the ith stimulus in the set of n stimuli. The sparseness has a maximal value of 1.0. It was found that the sparseness of the representation of the 68 stimuli by each neuron had an average across all neurons of 0.65. This indicates a rather distributed representation. 4. If the spontaneous firing rate was subtracted from the firing rate of the neuron to each stimulus, so that the changes of firing rate, i.e., the responses of the neurons, were used in the sparseness calculation, then the "response sparseness" had a lower value, with a mean of 0.33 for the population of neurons, or 0.60 if calculated over the set of faces. 5. Multidimensional scaling to produce a stimulus space represented by this population of neurons showed that the different faces were well separated in the space created, whereas the different nonface stimuli were grouped together in the space. 6. The information analyses and multidimensional scaling provided evidence that what was made explicit in the responses of these neurons was information about which face had been seen.(ABSTRACT TRUNCATED AT 400 WORDS)
Animals↗
Processing speed in the cerebral cortex and the neurophysiology of visual masking.
In experiments to investigate the duration of the time for which cortical neurons respond when the identification of a visual stimulus is just possible, we presented a test face stimulus for 16 ms, and followed it at different intervals by a masking stimulus (either an N-O pattern or a face) while recording from single neurons in the temporal visual cortex of macaques. When there was no mask the cells responded to the 16 ms of the test stimulus for 200-300 ms, far longer than the presentation time. We suggest that this reflects the operation of a short-term memory system implemented in cortical circuitry. If the mask was a stimulus which did not stimulate the cells (either a non-face pattern or a face which was a non-effective stimulus for that cell), then, as the interval between the onset of the test stimulus and the onset of the mask stimulus (the stimulus onset asynchrony) was reduced, the length of time for which the cells fired in response to the test stimulus was reduced. It is suggested that this is due to the mask stimulating adjacent cells in the cortex which by lateral inhibition reduce the responses of the cells activated by the test stimulus. When the stimulus onset asynchrony was 20 ms, face-selective neurons in the inferior temporal cortex of macaques responded for a period of 20-30 ms before their firing was interrupted by the mask. With the same test-mask stimulus onset asynchrony of 20 ms, humans could just identify which of six faces was shown.(ABSTRACT TRUNCATED AT 250 WORDS)
Animals↗
The responses of neurons in the temporal cortex of primates, and face identification and detection.
The ability of a human observer to detect the presence of a briefly flashed picture of a face can depend on the picture's spatial configuration, that is on whether its features are rearranged (jumbled) or are in their normal configuration. The face-detection effect (FDE) is found under conditions of backward masking, when the presence of a face can be detected with shorter masking intervals when it is in the normal than when in the rear-ranged configuration. A similar effect is found when the subject is asked to classify the face as rearranged or not - the face-classification effect (FCE). Part of the interest of the FDE and the FCE is that they show how the configuration of a stimulus can be an important factor in the perceptual processing which leads to detection and classification of the stimulus. To analyse these effects we recorded from single neurons in the cortex in the superior temporal sulcus of macaques when they were shown (in a visual fixation task) normal and rearranged faces under backward masking conditions shown in experiments 2 and 3 to produce, with the same apparatus, the FCE, and also to produce comparable effects on the identification of which face was present (called hereafter the face-identification effect), and also of the clarity of the face. We found in experiment 1 that there are some face-selective neurons which respond to faces only, or better, when the features in the faces are in their normal configuration rather than rearranged. We also showed in this experiment that the difference in the response to the normal as compared to the rearranged faces became greater when the masking stimulus was delayed more. Thus, at intermediate delays, there are more neurons active for the normal than for the rearranged face. We therefore propose that the FDE, the FCE, and the face-identification effect arise because the total number of neurons activated by faces in their normal configuration is greater than that activated by rearranged faces, because of the sensitivity of some face-selective neurons to the spatial arrangement of the features. The experiments also show that backward visual masking does produce abrupt termination of the firing of neurons in the temporal cortical visual system, so that the duration of a neuronal response is very short when visual stimuli can just be perceived.
Analysis of Variance↗
The molecular genetics and evolution of primate colour vision.
Until recently, the genetic basis of colour vision could only be inferred from measuring the colour vision of family groups. However, in the past few years the sites of the genes for visual pigments have been located and sequenced. The genes that specify the opsins for the rod and short-wavelength cone pigments are located on the third and seventh chromosomes, respectively. In Old World primates the genes for the middle- and long-wavelength pigments are located on the q arm of the X chromosome in a head-to-tail array. The close sequence similarity of the two genes on the X chromosome leads to a high frequency of unequal inter- and intragenic recombination leading to gene deletion or the creation of hybrid genes. In New World primates there is only a single locus on the X chromosome for a middle- to long-wavelength cone pigment. However, three alleles can occur at this locus and each codes for a slightly different cone pigment. As a result there are three types of male dichromat and three types of female dichromat and trichromat in each species. Colour vision in New World primates might be an intermediate stage between the uniform dichromacy of non-primate mammals and the uniform trichromacy of Old World primates. Alternatively, colour vision in New World primates might be an adaptation to allow a wide variety of colour-vision types within a single family group.
Alleles↗
Translation invariance in the responses to faces of single neurons in the temporal visual cortical areas of the alert macaque.
1. The responses of single neurons in the inferior temporal cortex and the cortex in the banks of the anterior part of the superior temporal sulcus of three awake, behaving macaques were recorded during a visual fixation task. Stimulus images subtending 17 or 8.5 degrees were presented in the center of the display area, and fixation was either at the center of the display area, or at one of four positions that were on the stimulus, or several degrees off the edge of the test stimulus. The experiments were performed with face-selective cells, and the responses were compared for fixation at each position for both effective and noneffective face stimuli for each cell. 2. The firing rates of most neurons to an effective image did not significantly alter when visual fixation was as far eccentric as the edge of the face, and they showed only a small reduction when the fixation point was up to 4 degrees from the edge of the face. Moreover, stimulus selectivity across faces was maintained throughout this region of the visual field. 3. The centers of the receptive fields of the cells, as shown by the calculated "centers of gravity," were close to the fovea, with almost all being within 3 degrees of the fovea. 4. The receptive fields of the cells typically crossed the vertical midline for at least 5 degrees. 5. Information theory procedures were used to analyze the spike trains of the visual neurons. Nearly six times more information was carried by these neurons' firing rate about the identity of an image than about its position in the visual field. Thus the information theory analysis showed that the responses of these neurons reflected information about which stimulus had been seen in a relatively translation invariant way. 6. Principal component analysis showed that principal component 1 (PC1) is related primarily to firing rate and reflected information primarily about stimulus identity. (For identity PC2 added only 14% more information to that contained in PC1.) Principal component 2 (PC2) was more closely related to neuronal response latencies, which increased with increasing eccentricity of the image in the visual field. PC2 reflected information about the position of the stimulus in the visual field, in that PC2 added 109% more information to that contained in PC1 about the position of the stimulus in the visual field.(ABSTRACT TRUNCATED AT 400 WORDS)
Animals↗
Oscillatory activity is not evident in the primate temporal visual cortex with static stimuli.
It has been suggested in studies in the visual system of anaesthetized cats that oscillatory activity with a frequency of 40-60 Hz occurs during the presentation of moving visual stimuli and reflects a synchronization process between neurons that could implement the binding together of related neurons into different sets. We found no evidence for such oscillations in the inferior temporal visual cortex and related areas of awake macaques fixating effective static visual stimuli, which for the neurons analysed were faces. We put forward the possibility that temporal synchronization between neurons to implement binding may not be generally used in the visual system as a solution to the binding problem, at least when static objects are being processed and recognised in higher parts of the visual system.
Animals↗
Representational capacity of face coding in monkeys.
We examine the distributed nature of the neural code for faces represented by the firing of visual neurons in the superior temporal sulcus of monkeys. Both information theory and neural decoding techniques are applied to determine how the capacity to represent faces depends on the number of coding neurons. Using a combination of experimental data and Monte Carlo simulations, we show that the information grows linearly and the capacity to encode stimuli grows exponentially with the number of neurons. By decoding firing rates, we determine that the responses of the 14 recorded neurons can distinguish between 20 face stimuli with approximately 80% accuracy. In general, we find that N neurons of this type can encode approximately 3(2(04N)) different faces with 50% discrimination accuracy. These results indicate that the neural code for faces is highly distributed and capable of accurately representing large numbers of stimuli.
Animals↗