Timothy J. Gawne

Unbiased measures of transmitted information and channel capacity from multivariate neuronal data

Two measures from information theory, transmitted information and channel capacity, can quantify the ability of neurons to convey stimulus-dependent information. These measures are calculated using probability functions estimated from stimulus-response data. However, these estimates are biased by response quantization, noise, and small sample sizes. Improved estimators are developed in this paper that depend on both an estimate of the sample-size bias and the noise in the data.

Attention influences single unit and local field potential response latencies in visual cortical area V4.

Many previous studies have demonstrated that changes in selective attention can alter the response magnitude of visual cortical neurons, but there has been little evidence for attention affecting response latency. Small latency differences, though hard to detect, can potentially be of functional importance, and may also give insight into the mechanisms of neuronal computation. We therefore reexamined the effect of attention on the response latency of both single units and the local field potential (LFP) in primate visual cortical area V4. We find that attention does produce small (1–2 ms) but significant reductions in the latency of both the spiking and LFP responses. Though attention, like contrast elevation, reduces response latencies, we find that the two have different effects on the magnitude of the LFP. Contrast elevations increase and attention decreases the magnitude of the initial deflection of the stimulus-evoked LFP. Both contrast elevation and attention increase the magnitude of the spiking response. We speculate that latencies may be reduced at higher contrast because stronger stimulus inputs drive neurons more rapidly to spiking threshold, while attention may reduce latencies by placing neurons in a more depolarized state closer to threshold before stimulus onset.

The Visual Evoked Potential is independent of surface alpha rhythm phase.

A Visual Evoked Potential (VEP) is an electrical signal picked up by a surface electrode in response to the activation of visual cortex by a visual stimulus. Because the VEP is typically much smaller in magnitude than the ongoing spontaneous EEG signal, the VEP is derived by averaging a large number of responses time-locked to stimulus presentation. Standard theory has it that the VEP is independent of the ongoing EEG, however, there has long been a competing view that the VEP is caused by a partial phase reset of the spontaneous alpha rhythm. We calculated the VEP where stimuli were presented at four different phases of the ongoing alpha rhythm, and subtracted away the responses to null trials synchronized to the same alpha rhythm phases, creating estimates of the VEP as a function of ongoing alpha rhythm phase. For some subjects there was evidence of an interaction between the VEP and the phase of the ongoing alpha rhythm, but this was idiosyncratic between subjects and conditions, and mostly evident in a later period when the VEP magnitude was very small. However, in general the VEP is independent of the phase of the ongoing alpha rhythm, and hence cannot be primarily caused by a partial phase resetting of the spontaneous EEG. It is possible that the VEP is either a phase-reset of an ongoing oscillation, or an oscillation induced by the sudden onset of a stimulus, but it cannot be the same oscillation as the surface alpha.

The simultaneous coding of orientation and contrast in the responses of V1 complex cells.

Abstract. The responses of 30 V1 complex cells were recorded using a complete set of transiently presented, oriented stimuli of different contrasts. A back-propagation neural network was used to predict the multivariate visual stimuli from the neuronal responses on a trial-by-trial basis. For single neurons, the strength of the response was much better at predicting the orientation of a visual stimulus than its contrast. Using the temporal modulation of the response improved the ability to predict the contrast of a stimulus without affecting the ability to predict the orientation. Removing stimulus latency from the responses, by time-shifting the individual responses an amount equal to the average latency, significantly reduced the ability to predict stimulus contrast, demonstrating that the response latency is reliable enough, even for a single neuron and a single trial, for it to be used to help determine stimulus contrast. Pooling the responses from a group of 11 neurons demonstrated that small groups of neurons can accurately represent multivariate stimuli in a single trial.

The relation between V1 neuronal responses and eye movement-like stimulus presentations

Abstract Primates normally make 2–3 saccadic eye movements/s to explore the environment. To investigate how these eye movements might influence visual responses, we compared the dynamics of stimuli arriving on V1 complex cell receptive fields by switching stimuli in sequence while a monkey fixated to the responses occurring when the stimulus appears due to saccadic eye movements. During the image sequences, information was greater when each image remained on the receptive fields longer, up to 200 ms; information was greatest when there was a gap of 50 ms between images. Responses were more variable when the image appeared due to a saccadic eye movement. The amount of stimulus-related information was lower in the early phase of the post-saccadic time, but increased during the post-saccadic fixation, so that after 400 ms there was almost as much stimulus-related information available as during the image switching. Eye position showed much larger variability after saccades, with the variability decreasing over 350–400ms to reach the level seen during long fixations. The dynamics of information accumulation in V1 complex cells appear to be well matched to the manner in which the environment is normally viewed.

Magnetic Transfer Contrast Accurately Localizes Substantia Nigra Confirmed by Histology

BACKGROUNDnMagnetic resonance imaging (MRI) has multiple contrast mechanisms. Like various staining techniques in histology, each contrast type reveals different information about the structure of the brain. However, it is not always clear how structures visible in MRI correspond to structures previously identified by histology. The purpose of this study was to determine if magnetic transfer contrast (MTC) or T2 contrast MRI was better at delineating the substantia nigra (SN).nnnMETHODSnMRI scans were acquired in vivo from two nonhuman primates (NHPs). The NHPs were subsequently euthanized, perfused, and their brains sectioned for histologic analyses. Each slice was photographed before sectioning. Each brain was sectioned into approximately 500 sections, 40 μm each, encompassing most of the cortex, midbrain, and dorsal parts of the hindbrain. Levels corresponding to anatomic MRI images were selected. From these, adjacent sections were stained using Kluver-Barrera (myelin and cell bodies) or tyrosine hydroxylase (dopaminergic neurons) immunohistochemistry. The resulting images were coregistered to the block-face images using a moving least squares algorithm with similarity transformations. MR images were similarly coregistered to the block-face images, allowing the structures on MRI to be identified with structures on the histologic images.nnnRESULTSnWe found that hyperintense (light) areas in MTC images were coextensive with the SN as delineated histologically. The hypointense (dark) areas in T2-weighted images were not coextensive with the SN but extended partially into the SN and partially into the cerebral peduncles.nnnCONCLUSIONSnMTC is more accurate than T2-weighting for localizing the SN in vivo.

The local and non-local components of the local field potential in awake primate visual cortex

The Local Field Potential (LFP) is the analog signal recorded from a microelectrode inserted into cortex, typically in the frequency band of approximately 1 to 200xa0Hz. Here visual stimuli were flashed on in the receptive fields of primary visual cortical neurons in awake behaving macaques, and both isolated single units (neurons) and the LFP signal were recorded from the same unipolar microelectrode. The fall-off of single unit activity as a visual stimulus was moved from near the center to near the edge of the receptive field paralleled the fall-off of the stimulus-locked (evoked) LFP response. This suggests that the evoked LFP strongly reflects local neuronal activity. However, the evoked LFP could be significant even when the visual stimulus was completely outside the receptive field and the single unit response had fallen to zero, although this phenomenon was variable. Some of the non-local components of the LFP may be related to the slow distributed, or non-retinotopic, LFP signal previously observed in anesthetized animals. The induced (not time-locked to stimulus onset) component of the LFP showed significant increases only for stimuli within the receptive field of the single units. While the LFP primarily reflects local neuronal activity, it can also reflect neuronal activity at more distant sites, although these non-local components are typically more variable, slower, and weaker than the local components.

Neuronal codes: reading them and learning how their structure influences network organization

Our investigations of the primate visual system show that neuronal responses carry information in a multi-dimensional code that is superimposed onto the response envelope in a slow time varying fashion. The precision of timing is 30 ms or more. In primary visual cortex response latency and response strength are largely independent, with latency more closely coding contrast or visibility and strength more closely coding stimulus orientation, or perhaps shape. Adjacent neurons in both V1 and inferior temporal cortex share only about 10% of their stimulus-related information, which we demonstrate to be consistent with the idea that cortical layers were organized to minimize information loss.

Visual acuity of the midland banded water snake estimated from evoked telencephalic potentials

The visual acuity of seven midland banded water snakes was measured by recording evoked responses from telencephalon to temporally modulated square wave grating patterns. Using conventional electrophysiological techniques and signal averaging, high contrast square wave gratings of different spatial frequencies were presented. Acuity was estimated by extrapolating relative response amplitude/log10 spatial frequency functions which yielded an average acuity of 4.25xa0cycles/degree. Refractive state was also estimated by recording evoked potentials to intermediate spatial frequencies with different lenses in front of the eye. Polynomial fits indicated that under the experimental conditions the snakes were around 6.4 diopters hyperopic suggesting a corrected acuity of 4.89xa0cycles/degree. Reduction of grating luminance resulted in a reduction in evoked potential acuity measurements. These results indicate that the spatial resolution of midland banded water snakes is the equal of cat; about 20/120 in human clinical terms.

The wavelength composition and temporal modulation of ambient lighting strongly affect refractive development in young tree shrews

Abstract Shortly after birth, the eyes of most animals (including humans) are hyperopic because the short axial length places the retina in front of the focal plane. During postnatal development, an emmetropization mechanism uses cues related to refractive error to modulate the growth of the eye, moving the retina toward the focal plane. One possible cue may be longitudinal chromatic aberration (LCA), to signal if eyes are getting too long (long [red] wavelengths in better focus than short [blue]) or too short (short wavelengths in better focus). It could be difficult for the short‐wavelength sensitive (SWS, “blue”) cones, which are scarce and widely spaced across the retina, to detect and signal defocus of short wavelengths. We hypothesized that the SWS cone retinal pathway could instead utilize temporal (flicker) information. We thus tested if exposure solely to long‐wavelength light would cause developing eyes to slow their axial growth and remain refractively hyperopic, and if flickering short‐wavelength light would cause eyes to accelerate their axial growth and become myopic. Four groups of infant northern tree shrews (Tupaia glis belangeri, dichromatic mammals closely related to primates) began 13 days of wavelength treatment starting at 11 days of visual experience (DVE). Ambient lighting was provided by an array of either long‐wavelength (red, 626 ± 10 nm) or short‐wavelength (blue, 464 ± 10 nm) light‐emitting diodes placed atop the cage. The lights were either steady, or flickering in a pseudo‐random step pattern. The approximate mean illuminance (in human lux) on the cage floor was red (steady, 527 lux; flickering, 329 lux), and blue (steady, 601 lux; flickering, 252 lux). Refractive state and ocular component dimensions were measured and compared with a group of age‐matched normal animals (n = 15 for refraction (first and last days); 7 for ocular components) raised in broad spectrum white fluorescent colony lighting (100–300 lux). During the 13 day period, the refraction of the normal animals decreased from (mean ± SEM) 5.8 ± 0.7 diopters (D) to 1.5 ± 0.2 D as their vitreous chamber depth increased from 2.77 ± 0.01 mm to 2.80 ± 0.03 mm. Animals exposed to red light (both steady and flickering) remained hyperopic throughout the treatment period so that the eyes at the end of wavelength treatment were significantly hyperopic (7.0 ± 0.7 D, steady; 4.7 ± 0.8 D, flickering) compared with the normal animals (p < 0.01). The vitreous chamber of the steady red group (2.65 ± 0.03 mm) was significantly shorter than normal (p < 0.01). On average, steady blue light had little effect; the refractions paralleled the normal refractive decrease. In contrast, animals housed in flickering blue light increased the rate of refractive decrease so that the eyes became significantly myopic (−2.9 ± 1.3 D) compared with the normal eyes and had longer vitreous chambers (2.93 ± 0.04 mm). Upon return to colony lighting, refractions in all groups gradually returned toward emmetropia. These data are consistent both with the hypothesis that LCA can be an important visual cue for postnatal refractive development, and that short‐wavelength temporal flicker provides an important cue for assessing and signaling defocus. Graphical abstract Figure. No Caption available. HighlightsExamined effect of long and short wavelengths on refractive development.Studied tree shrews, cone‐dominated dichromatic mammals closely related to primates.Narrow‐band red light (626 nm) slowed vitreous chamber growth, producing hyperopia.Flickering narrow‐band blue light (464 nm) produced elongated vitreous and myopia.Wavelength exposure in young tree shrews has powerful effects on emmetropization.