Detection of sinusoidal gratings by pattern-specific detectors: further evidence for the correlation principle in human vision

 The detection of compound sinusoidal gratings of various spatial frequency separations and four different grating sizes has been studied using the summation-to-threshold paradigm. Contrast interrelation functions have been measured and spatial frequency tuning estimates, based on the slope of the contrast interrelation function at two definite points, were derived using the “negative gradient technique” proposed by Logvinenko [Logvinenko (1995) Biol Cybern 73: 547–552]. It is shown that compound grating detection can be modelled by assuming pattern-specific sensory mechanisms for each of the spatial frequency components, which adapt to the periodicity and the size of the stimulus but not to its envelope function. Further, it is shown that relative sensitivity for a given spatial frequency separation can be predicted with good accuracy by the correlation of the grating components used for superposition. It is suggested that the most plausible implementation of the pattern correlation principle in human grating detection is the “grating cell” model. Received: 10 March 2000 / Accepted in revised form: 4 December 2000     

Discharge patterns of chicken cochlear ganglion neurons following kanamycin-induced hair cell loss and regeneration

Hair cells in the basal, high frequency region (>1100 Hz) of the chicken cochlea were destroyed with kanamycin (400 mg/kg/d × 10 d) and allowed to regenerate. Afterwards, single unit recordings were made from cochlear ganglion neurons at various times post-treatment. During the first few weeks post-treatment, only neurons with low characteristic frequencies (<1100 Hz) responded to sound. Despite the fact that the low frequency region of the cochlea was not destroyed, neurons with low characteristic frequencies had elevated thresholds, abnormally broad U-shaped or W-shaped tuning curves and low spontaneous discharge rates. At 2 days post-treatment, the spontaneous discharge rates of some acoustically unresponsive units fluctuated in a rhythmical manner. As recovery time increased, thresholds decreased, tuning curves narrowed and developed a symmetrical V-shape, spontaneous rate increased and neurons with higher characteristic frequencies began to respond to sound. In addition, the proportion of interspike interval histograms with regularly spaced peaks increased. These improvements progressed along a low-to-high characteristic frequency gradient. By 10–20 weeks post-treatment, the thresholds and tuning curves of neurons with characteristic frequencies below 2000 Hz were within normal limits; however, the spontaneous discharge rates of the neurons were still significantly lower than those from normal animals.Abbreviations KM kanamycin - BrdU bromodeoxyuridine - CF characteristic frequency - CAP compound action potential - ISI interspike interval     

猫外膝体神经元非寻常的方位调谐特性——Ⅰ、在正常猫的研究

在九只成年猫上用玻璃电极记录了单个外膝体神经元对不同方位的移动正弦光栅刺激的反应共详细测定了４００个细胞的方位调谐特性。少数外膝体神经元具有非寻常的方位调谐特性,包括：具蝴蝶状调谐曲线的方位调谐特性;双调谐的方位调谐特性和最优方位随刺激光栅空间频率的改变而变化的方位调谐特性。这些细胞非寻常的方位调谐特性往往伴随着非寻常的空间频率调谐特性。空们的方位调谐特性和空间频率调谐特性都不能用Ｓｏｏｄａｋ等提     

Different neuroplasticity for task targets and distractors

Adult learning-induced sensory cortex plasticity results in enhanced action potential rates in neurons that have the most relevant information for the task, or those that respond strongly to one sensory stimulus but weakly to its comparison stimulus. Current theories suggest this plasticity is caused when target stimulus evoked activity is enhanced by reward signals from neuromodulatory nuclei. Prior work has found evidence suggestive of nonselective enhancement of neural responses, and suppression of responses to task distractors, but the differences in these effects between detection and discrimination have not been directly tested. Using cortical implants, we defined physiological responses in macaque somatosensory cortex during serial, matched, detection and discrimination tasks. Nonselective increases in neural responsiveness were observed during detection learning. Suppression of responses to task distractors was observed during discrimination learning, and this suppression was specific to cortical locations that sampled responses to the task distractor before learning. Changes in receptive field size were measured as the area of skin that had a significant response to a constant magnitude stimulus, and these areal changes paralleled changes in responsiveness. From before detection learning until after discrimination learning, the enduring changes were selective suppression of cortical locations responsive to task distractors, and nonselective enhancement of responsiveness at cortical locations selective for target and control skin sites. A comparison of observations in prior studies with the observed plasticity effects suggests that the non-selective response enhancement and selective suppression suffice to explain known plasticity phenomena in simple spatial tasks. This work suggests that differential responsiveness to task targets and distractors in primary sensory cortex for a simple spatial detection and discrimination task arise from nonselective increases in response over a broad cortical locus that includes the representation of the task target, and selective suppression of responses to the task distractor within this locus.     

Modeling complex tone perception: grouping harmonics with combination-sensitive neurons

 Perception of complex communication sounds is a major function of the auditory system. To create a coherent percept of these sounds the auditory system may instantaneously group or bind multiple harmonics within complex sounds. This perception strategy simplifies further processing of complex sounds and facilitates their meaningful integration with other sensory inputs. Based on experimental data and a realistic model, we propose that associative learning of combinations of harmonic frequencies and nonlinear facilitation of responses to those combinations, also referred to as “combination-sensitivity,” are important for spectral grouping. For our model, we simulated combination sensitivity using Hebbian and associative types of synaptic plasticity in auditory neurons. We also provided a parallel tonotopic input that converges and diverges within the network. Neurons in higher-order layers of the network exhibited an emergent property of multifrequency tuning that is consistent with experimental findings. Furthermore, this network had the capacity to “recognize” the pitch or fundamental frequency of a harmonic tone complex even when the fundamental frequency itself was missing. Received: 6 October 2001 / Accepted in revised form: 21 January 2002     

Relation of spatial and temporal frequencies to vernier acuity

Vernier thresholds were measured with a pair of vertical sinusoidal gratings of one and a half cycles as targets. The amplitude was weighted by a one-dimensional Gaussian and contrast was set one log unit above contrast threshold. The vernier thresholds were estimated with the method of constant stimuli. Temporal frequency effects were introduced by movement of the vernier targets. It was found that vernier thresholds expressed in phase angle were unchanged in the effective range of spatial frequencies provided that the temporal frequency and the visibility were unchanged, and that thresholds deteriorated by increasing the temporal frequency. It is suggested that the detection of relative phase may be involved in the discrimination of vernier offsets and that it may be mediated by a sustained unit. Three possible types of mechanisms, edge-localization processes, orientation-selective units and phase-sensitive units, were considered in relation to vernier acuity.     

The oscar, <Emphasis Type="Italic">Astronotus ocellatus</Emphasis>, detects and discriminates dipole stimuli with the lateral line system

We studied the role of the lateral line system for detection and discrimination of dipole stimuli in the oscar, Astronotus ocellatus (Family Cichlidae), and determined detection thresholds in still water and frequency discrimination capabilities in still and turbulent water. Average detection threshold of six animals for a 100-Hz dipole stimulus was 0.0059 μm peak-to-peak water displacement at the surface of the fish. After inactivation of the neuromast receptor organs of the lateral line system with the antibiotic streptomycin, dipole detection was reduced, but recovered within 2–4 weeks. This suggests that the oscar relied strongly on hydrodynamic information received by the lateral line system. Five oscars learned to discriminate a 100-Hz stimulus from 70 Hz and lower frequencies. When turbulence was introduced into the experimental tank, fish were still able to discriminate 100 Hz from frequencies 70 Hz and lower indicating that frequency discrimination mediated by the lateral line system was not reduced in turbulent water.     

A computational model of rapid task-related plasticity of auditory cortical receptive fields

Receptive field properties of neurons in A1 can rapidly adapt their shapes during task performance in accord with specific task demands and salient sensory cues (Fritz et al., Hearing Research, 206:159–176, 2005a, Nature Neuroscience, 6: 1216–1223, 2003). Such modulatory changes selectively enhance overall cortical responsiveness to target (foreground) sounds and thus increase the likelihood of detection against the background of reference sounds. In this study, we develop a mathematical model to describe how enhancing discrimination between two arbitrary classes of sounds can lead to the observed receptive field changes in a variety of spectral and temporal discrimination tasks. Cortical receptive fields are modeled as filters that change their spectro-temporal tuning properties so as to respond best to the discriminatory acoustic features between foreground and background stimuli. We also illustrate how biologically plausible constraints on the spectro-temporal tuning of the receptive fields can be used to optimize the plasticity. Results of the model simulations are compared to published data from a variety of experimental paradigms.     

Absolute hearing thresholds and critical masking ratios in the European barn owl: a comparison with other owls

Absolute thresholds and critical masking ratios were determined behaviorally for the European barn owl (Tyto alba guttata). It shows an excellent sensitivity throughout its hearing range with a minimum threshold of −14.2 dB sound pressure level at 6.3 kHz, which is similar to the sensitivity found in the American barn owl (Tyto alba pratincola) and some other owls. Both the European and the American barn owl have a high upper-frequency limit of hearing exceeding that in other bird species. Critical masking ratios, that can provide an estimate for the frequency selectivity in the barn owl's hearing system, were determined with a noise of about 0 dB spectrum level. They increased from 19.1 dB at 2 kHz to 29.2 dB at 8 kHz at a rate of 5.1 dB per octave. The corresponding critical ratio bandwidths were 81, 218, 562 and 831 Hz for test-tone frequencies of 2, 4, 6.3 and 8 kHz, respectively. These values indicate, contrary to expectations based on the spatial representation of frequencies on the basilar papilla, increasing bandwidths of auditory filters in the region of the barn owl's auditory fovea. This increase, however, correlates with the increase in the bandwidths of tuning curves in the barn owl's auditory fovea. Accepted: 27 November 1997     

Acoustic distortion products from the cochlea of the blind African mole rat, Cryptomys spec.

The measurement of distortion-product otoacoustic emissions is a noninvasive method that can be used for assessing the sensitivity and the frequency tuning of nonlinear cochlear mechanics. During stimulation with two pure tones f1 and f2, the acoustic 2f1-f2 distortion was recorded in the ear canal of Cryptomys spec. to study specializations in cochlear mechanics that could be associated with the presence of a frequency expanded cochlear region between 0.8–1 kHz. In addition, a distortion threshold curve was obtained which describes relative threshold of nonlinear cochlear mechanics. Sensitive distortion thresholds could be measured for stimulus frequencies between 0.4 to 18 kHz with a broad minimum between 0.75 to 2.5 kHz. The distortion threshold curve extends to higher frequencies than previous neuronal data indicated.As a measure of mechanical tuning sharpness in the cochlea, suppression tuning curves of 2f1-f2 were recorded. The tuning curves reflected the typical mammalian pattern with shallow low frequency and steep high frequency slopes. Their tuning sharpness was poor with Q10dB values between 0.3 and 1.88. In the range of the frequency expanded region, the Q10dB values were below 0.5. This finding emphasizes that the presence of frequency expansion does not necessarily lead to enhanced mechanical tuning in the cochlea and one has to consider if in certain bat species with cochlear frequency expansion and particularly sharp cochlear tuning, the two phenomena may not be interlinked.Abbreviations CF constant frequency component of echolocation call - STC suppression tuning curve     

The Opponent Channel Population Code of Sound Location Is an Efficient Representation of Natural Binaural Sounds

In mammalian auditory cortex, sound source position is represented by a population of broadly tuned neurons whose firing is modulated by sounds located at all positions surrounding the animal. Peaks of their tuning curves are concentrated at lateral position, while their slopes are steepest at the interaural midline, allowing for the maximum localization accuracy in that area. These experimental observations contradict initial assumptions that the auditory space is represented as a topographic cortical map. It has been suggested that a “panoramic” code has evolved to match specific demands of the sound localization task. This work provides evidence suggesting that properties of spatial auditory neurons identified experimentally follow from a general design principle- learning a sparse, efficient representation of natural stimuli. Natural binaural sounds were recorded and served as input to a hierarchical sparse-coding model. In the first layer, left and right ear sounds were separately encoded by a population of complex-valued basis functions which separated phase and amplitude. Both parameters are known to carry information relevant for spatial hearing. Monaural input converged in the second layer, which learned a joint representation of amplitude and interaural phase difference. Spatial selectivity of each second-layer unit was measured by exposing the model to natural sound sources recorded at different positions. Obtained tuning curves match well tuning characteristics of neurons in the mammalian auditory cortex. This study connects neuronal coding of the auditory space with natural stimulus statistics and generates new experimental predictions. Moreover, results presented here suggest that cortical regions with seemingly different functions may implement the same computational strategy-efficient coding.     

Echo SPL, training experience, and experimental procedure influence the ranging performance in the big brown bat, Eptesicus fuscus

Four Eptesicus fuscus were trained in a range discrimination experiment to choose the closer of two phantom targets. Echo attenuation was roving between trials returning echoes ranging from −10 dB to −50 dB SPL (sound pressure level) relative to emission SPL. Discrimination thresholds were determined. After sufficient training, ranging performance was stable and about the same in the range between −20 dB and −50 dB with range difference thresholds around 300 μs. At −10 dB, performance was poor even after long training. After additional training at a constant relative echo SPL of −30 dB and a delay difference of 300 μs the performance measured with roving echo SPL improved at all relative echo SPL between −20 dB and −50 dB but not at −10 dB. The new experimental procedure improved the performance by additional learning, and the bats generalized over a wide range of relative echo SPL. Threshold improved to 100 μs when measured at a constant relative echo SPL of −30 dB, again indicating the influence of the experimental procedure. In correspondence to neurophysiological data the ranging performance deteriorates if the echo SPL is close to the emission SPL. Signal duration and emission SPL were variable during range discrimination. Accepted: 7 March 1998     

Pupillary dilation response as an indicator of auditory discrimination in the barn owl

The pupil of an awake, untrained, head-restrained barn owl was found to dilate in response to sounds with a latency of about 25 ms. The magnitude of the dilation scaled with signal-to-noise ratio. The dilation response habituated when a sound was repeated, but recovered when stimulus frequency or location was changed. The magnitude of the recovered response was related to the degree to which habituating and novel stimuli differed and was therefore exploited to measure frequency and spatial discrimination. Frequency discrimination was examined by habituating the response to a reference tone at 3 kHz or 6 kHz and determining the minimum change in frequency required to induce recovery. We observed frequency discrimination of 125 Hz at 3 kHz and 250 Hz at 6 kHz – values comparable to those reported by others using an operant task. Spatial discrimination was assessed by habituating the response to a stimulus from one location and determining the minimum horizontal speaker separation required for recovery. This yielded the first measure of the minimum audible angle in the barn owl: 3° for broadband noise and 4.5° for narrowband noise. The acoustically evoked pupillary dilation is thus a promising indicator of auditory discrimination requiring neither training nor aversive stimuli. Accepted: 28 February 2000     

Spike-based synaptic plasticity and the emergence of direction selective simple cells: mathematical analysis

In the companion paper we presented extended simulations showing that the recently observed spike-timing dependent synaptic plasticity can explain the development of simple cell direction selectivity (DS) when simultaneously modifying the synaptic strength and the degree of synaptic depression. Here we estimate the spatial shift of the simple cell receptive field (RF) induced by the long-term synaptic plasticity, and the temporal phase advance caused by the short-term synaptic depression in response to drifting grating stimuli. The analytical expressions for this spatial shift and temporal phase advance lead to a qualitative reproduction of the frequency tuning curves of non-directional and directional simple cells. In agreement with in vivo recordings, the acquired DS is strongest for test gratings with a temporal frequency around 1–4 Hz. In our model this best frequency is determined by the width of the learning function and the time course of depression, but not by the temporal frequency of the training stimuli. The analysis further reveals the instability of the initially symmetric RF, and formally explains why direction selectivity develops from a non-directional cell in a natural, directionally unbiased stimulation scenario.     

Odor detection thresholds in the rat for the vapors of three related perfluorocarbons and ethylene glycol dinitrate*

As a first step in ascertaining the feasibility of using laboratoryrats for explosive detection, we evaluated their sensitivityto vapor-phase concentrations of the explosive ethylene glycoldinitrate (EGDN) and three homologous perfluorocarbons, perfluoromethylcyciohexane(PMCH) perfluorodimethylcyclohexane (PDCH), and perfluorodecalin(PFD). These perfluorocarbons are under evaluation by the U.S.Government as potential "explosive taggants" – substancesadded to explosives which identify their source of manufactureor distribution. The mean detection thresholds of these compounds,as determined in an operant discrimination task using waterreinforcement, were as follows: EGDN, 0.05 ppm (range 0.05–0.06);PDCH, 1.4 ppm (range 0.60–6.50); PMCH, 2.1 ppm (range1.85–4.65); and PFD, 1.1 ppm (range 0.60–1.69).This study provides the first quantitative data on the olfactorysensitivity of rats to the vapors of explosive/taggant compoundsand suggests that the detection curves for such odorants arefundamentally similar to those of other stimuli that have beenevaluated in previous work.     

Sensory coding and contrast invariance emerge from the control of plastic inhibition over emergent selectivity

Visual stimuli are represented by a highly efficient code in the primary visual cortex, but the development of this code is still unclear. Two distinct factors control coding efficiency: Representational efficiency, which is determined by neuronal tuning diversity, and metabolic efficiency, which is influenced by neuronal gain. How these determinants of coding efficiency are shaped during development, supported by excitatory and inhibitory plasticity, is only partially understood. We investigate a fully plastic spiking network of the primary visual cortex, building on phenomenological plasticity rules. Our results suggest that inhibitory plasticity is key to the emergence of tuning diversity and accurate input encoding. We show that inhibitory feedback (random and specific) increases the metabolic efficiency by implementing a gain control mechanism. Interestingly, this led to the spontaneous emergence of contrast-invariant tuning curves. Our findings highlight that (1) interneuron plasticity is key to the development of tuning diversity and (2) that efficient sensory representations are an emergent property of the resulting network.     

The Sign Rule and Beyond: Boundary Effects,Flexibility, and Noise Correlations in Neural Population Codes

Over repeat presentations of the same stimulus, sensory neurons show variable responses. This “noise” is typically correlated between pairs of cells, and a question with rich history in neuroscience is how these noise correlations impact the population''s ability to encode the stimulus. Here, we consider a very general setting for population coding, investigating how information varies as a function of noise correlations, with all other aspects of the problem – neural tuning curves, etc. – held fixed. This work yields unifying insights into the role of noise correlations. These are summarized in the form of theorems, and illustrated with numerical examples involving neurons with diverse tuning curves. Our main contributions are as follows. (1) We generalize previous results to prove a sign rule (SR) — if noise correlations between pairs of neurons have opposite signs vs. their signal correlations, then coding performance will improve compared to the independent case. This holds for three different metrics of coding performance, and for arbitrary tuning curves and levels of heterogeneity. This generality is true for our other results as well. (2) As also pointed out in the literature, the SR does not provide a necessary condition for good coding. We show that a diverse set of correlation structures can improve coding. Many of these violate the SR, as do experimentally observed correlations. There is structure to this diversity: we prove that the optimal correlation structures must lie on boundaries of the possible set of noise correlations. (3) We provide a novel set of necessary and sufficient conditions, under which the coding performance (in the presence of noise) will be as good as it would be if there were no noise present at all.     

Frequency-intensity characteristics of cricket cercal interneurons: units with high-pass functions

Interneurons in the cercal sensory system of crickets respond in a cell-specific manner if the cercal hair sensilla are stimulated by air-particle oscillations at frequencies below about 2000 Hz. We investigated the filter properties of several of these interneurons, and tested the effect of stimulus intensity (typically 0.3–50 mm s⁻¹ peak-to-peak air-particle velocity) on the frequency response in the range 5–600 Hz. We focus on three interneurons (the lateral and medial giant interneurons and interneuron 9-3a) of Acheta domesticus which are characterized by a relatively high sensitivity above ca. 50–200 Hz. The responses of the medial giant interneuron usually increase monotonically with frequency and intensity. Interneuron 9-3a and the lateral giant interneuron exhibit saturation or response decrement at high frequencies and intensities. The lateral giant interneuron has an additional peak of sensitivity below about 40 Hz. Small individual variations in the relative locations of the two response areas of this interneuron within the frequency-intensity field are responsible for a large variability obtained if frequency-response curves are determined for particular intensities. Stimulus frequency does not affect the principal directional preferences of the three interneurons. Nevertheless, if tested individually, the lateral giant interneuron and interneuron 9-3a exhibit small changes of directional tuning. Accepted: 12 November 1997     

Auditory Discrimination Learning: Role of Working Memory

Perceptual training is generally assumed to improve perception by modifying the encoding or decoding of sensory information. However, this assumption is incompatible with recent demonstrations that transfer of learning can be enhanced by across-trial variation of training stimuli or task. Here we present three lines of evidence from healthy adults in support of the idea that the enhanced transfer of auditory discrimination learning is mediated by working memory (WM). First, the ability to discriminate small differences in tone frequency or duration was correlated with WM measured with a tone n-back task. Second, training frequency discrimination around a variable frequency transferred to and from WM learning, but training around a fixed frequency did not. The transfer of learning in both directions was correlated with a reduction of the influence of stimulus variation in the discrimination task, linking WM and its improvement to across-trial stimulus interaction in auditory discrimination. Third, while WM training transferred broadly to other WM and auditory discrimination tasks, variable-frequency training on duration discrimination did not improve WM, indicating that stimulus variation challenges and trains WM only if the task demands stimulus updating in the varied dimension. The results provide empirical evidence as well as a theoretic framework for interactions between cognitive and sensory plasticity during perceptual experience.     

Stimulus sampling as an exploration mechanism for fast reinforcement learning

Reinforcement learning in neural networks requires a mechanism for exploring new network states in response to a single, nonspecific reward signal. Existing models have introduced synaptic or neuronal noise to drive this exploration. However, those types of noise tend to almost average out—precluding or significantly hindering learning —when coding in neuronal populations or by mean firing rates is considered. Furthermore, careful tuning is required to find the elusive balance between the often conflicting demands of speed and reliability of learning. Here we show that there is in fact no need to rely on intrinsic noise. Instead, ongoing synaptic plasticity triggered by the naturally occurring online sampling of a stimulus out of an entire stimulus set produces enough fluctuations in the synaptic efficacies for successful learning. By combining stimulus sampling with reward attenuation, we demonstrate that a simple Hebbian-like learning rule yields the performance that is very close to that of primates on visuomotor association tasks. In contrast, learning rules based on intrinsic noise (node and weight perturbation) are markedly slower. Furthermore, the performance advantage of our approach persists for more complex tasks and network architectures. We suggest that stimulus sampling and reward attenuation are two key components of a framework by which any single-cell supervised learning rule can be converted into a reinforcement learning rule for networks without requiring any intrinsic noise source. This work was supported by the Swiss National Science Foundation grant K-32K0-118084.