GIOVANNE ROSSO - Federal University of Rio Grande do … · universidade federal do rio grande do norte! programa de pÓs-graduaÇÃo em neurociÊncias! "! !!! do fast retinal oscillations

UNIVERS IDADE FEDERAL DO RIO GRANDE DO NORTE

PROGRAMA DE PÓS-GRADUAÇÃO EM NEUROCIÊNCIAS

� !!!!

DO FAST RET INAL OSC ILLAT IONS PLAY A ROLE IN V I S ION? A STUDY IN THE ANESTHET IZED AND AWAKE C AT

!!

G I O V A N N E R O S S O

!!!!!!!!!

N a t a l , 2 0 1 5

I N S T I T U TO D O !C É R E B RO

!UNIVERS IDADE FEDERAL DO RIO GRANDE DO NORTE


�� !!!!


!!!


!TRABALHO APRESENTADO AO PROGRAMA DE PÓS-GRADUAÇÃO EM NEUROCIÊNCIAS DA UNIVERSIDADE

FEDERAL DO RIO GRANDE DO NORTE COMO REQUISITO PARCIAL PARA A OBTENÇÃO DO

G R A U D E M E S T R E !OR I E N TA D O R : Pro f . Dr. SERGIO NEUENSCHWANDER

NEUROBIOLOGIA DE SISTEMAS E COGNIÇÃO

!!!!

N a t a l , 2 0 1 5


UNIVERS IDADE FEDERAL DO RIO GRANDE DO NORTE


�� !!!


!!!


!DISSERTAÇÃO APRESENTADA AO PROGRAMA DE PÓS-GRADUAÇÃO EM NEUROCIÊNCIAS

DA UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE

COMO REQUISITO PARCIAL PARA A OBTENÇÃO DO GRAU DE MESTRE.

ÁREA DE CONCENTRAÇÃO: NEUROBIOLOGIA DE SISTEMAS E COGNIÇÃO. !!APROVADA EM: 16.11.2015

B A N C A E X A M I N A D O R A : PROF. DR. SERGIO NEUENSCHWANDER

PROF. DR. CLAUDIO QUEIROZ

PROF. DR. JEROME BARON


!!!!!!!!!!!!!!

!EXPERIMENTS ARE THE ONLY MEANS OF KNOWLEDGE AT OUR DISPOSAL.

THE REST IS POETRY, IMAGINATION.

MAX PLANCK !

AB S T R A C T

Early physiologists were dazzled by the occurrence of high-amplitude, periodic oscillations, easily discernible in recording traces from the eye, optic tract and optic ganglia. Numerous studies thereafter pointed to retinal ganglion cell as the elements responsible for the generation of these fast rhythms, which were known to propagate to the lateral geniculate and to the cortex. Only recently, however, these early observations gained renewed interest, mainly in the light of recent theories linking neuronal oscillations to various cognitive processes, such as perceptual binding, attention and memory. In this context, fast retinal oscillations have been associated to the binding of contiguous contours or surfaces, which in principle could support a fast feedforward segmentation process. In addition, a series of experiments in the cat have shown that fast oscillations in the retina may convey global stimulus properties, such as size.

A limitation in these previous studies, however, was that most of them where were made in the anesthetized and paralyzed cat. Only a few early studies have been performed in the non-anesthetized but still paralyzed cat. Another concern was that, in these latter experiments, visual stimuli were often limited to ganzfeld flashes, far from natural vision conditions. Moreover, very recently we made the surprising observation that fast retinal oscillations depend strongly on halothane (and isoflurane) anesthesia. It was therefore imperative to verify whether oscillatory activity is also present in the awake cat, under naturalistic conditions, such as during free-viewing of a visual scene. This is the main goal of the present study.

Simultaneous multiple-electrode recordings were made from the lateral geniculate nucleus (LGN) and the retina of anesthetized cats (N= 3) and from the LGN of an awake cat (N= 1). Comparisons were made for responses to natural movies and flashed stationary light stimuli. To test specifically the role of retinal oscillations in encoding stimulus size we designed a protocol made of a light circle of varying size along the trial. Spike sorting techniques allowed us to study separately the ON- and OFF-components of the responses. Analysis consisted in measuring synchronous oscillations for single cell spiking activity in the time (sliding correlation analysis) and spectral domains (multitaper spectral analysis, multitaper coherence). Our present results based on single-cells extend our previous findings in the anesthetized cat, which were restricted to an autocorrelation analysis of LGN mutiunitary responses. Both ON- and OFF-responses to varying size stimuli show that coherent oscillations appear only after the stimulus attained a minimum size of about 5° (depending on the contrast level), suggesting that oscillations in the retina are rather limited in encoding subtle changes in stimulus size. Recordings obtained directly from eye showed that oscillations in the retina, as in the LGN, are highly correlated with the concentrations level of halothane. Notably, in a series of sessions we were able to record LGN responses in an awake cat, which was subsequently anesthetized with halothane, keeping the same recording site. Oscillations were completely absent in the awake condition and appeared strong as usual during the halothane anesthesia.

Overall these results weaken substantially the notion that fast retinal oscillations are meaningful for vision. Nevertheless, as shown from our single cell analysis, retinal oscillations share many of the properties of cortical gamma oscillations. In this respect, oscillations in the retina induced by halothane serve as a valuable preparation, even though artificial, for studying oscillatory neuronal dynamics. !!

Key words: retina, geniculate, oscillation, coherence, halothane, awake.

RE S U M O

Os primeiros fisiologistas ficaram certamente impressionados com a existência de oscilações periódicas de alta amplitude, claramente visíveis nos traçados obtidos da retina, trato óptico e gânglios ópticos. Posteriormente vários estudos mostraram ser a células ganglionares os elementos responsáveis pela geração destes ritmos rápidos, que sabia-se podem propagar da retina ao geniculado lateral e ao córtex. Apenas recentemente, no entanto, estas observações ganharam novo interesse, principalmente a luz de teorias e conjecturas que atribuem às oscilações neuronais vários processos cognitivos, como a ligação perceptual, a atenção e a memória. Segundo esta hipótese, oscilações rápidas da retina seriam importantes para a ligação de contornos contíguos ou superfícies, podendo assim constituir um mecanismo feedforward importante na segmentação visual. Em acordo com estas noções, uma série de experimentos no gato mostraram que oscilações rápidas da retina podem ser informativas sobre propriedades globais do estímulo como o seu tamanho.

Uma grande limitação nestes estudos, no entanto, foi o fato de terem sido feitos sob anestesia e paralisia. Apenas alguns experimentos foram realizados em gatos não-anestesiados, mesmo assim, paralisados. Uma outra limitação foi o uso de estímulos visuais limitados a breves exposições, que ocupavam todo o campo visual, muito longe de condições naturais da visão. Por outro lado, muito recentemente, fizemos uma observação inesperada no nosso laboratório: oscilações rápidas da retina dependem fortemente da anestesia por halotano (e isoflurano). Tornou-se assim imperativo investigar se as oscilações rápidas da retina estão presentes ou não no gato não anestesiado, em condições naturais, como por exemplo durante a observação-livre de uma cena visual. Este é o principal objetivo deste estudo.

Para isto, registros simultâneos através de eletródios-múltiplos foram feitos no geniculado lateral e na retina de gatos anestesiados (N= 3) e acordado (N= 1). Comparações foram feitas para respostas a filmes de cenas naturais e estímulos estacionários, como círculos luminosos. Para testar especificamente o papel das oscilações rápidas da retina na codificação do tamanho do estímulo visual aplicamos um protocolo que consiste em apresentar sobre os campos receptores um círculo luminoso de tamanho variável ao longo do tempo. Técnicas de separação de potenciais-de-ação nos permitiu estudar individualmente os componentes ON e OFF das respostas multi-unitárias. Nossa análise consistiu em obter medidas das oscilações sincrônicas para células isoladas ao longo do tempo no domínio temporal (análise de correlação por janela deslizante) e no domínio espectral (análise espectral por afunilamento múltiplo, coerência por afunilamento múltiplo). Estes resultados estendem os nossos achados prévios no gato anestesiado, que foram restritos à análise de auto-correlação de repostas multi-unitárias do geniculado lateral. Tanto as repostas ON como as respostas OFF a estímulos visuais de tamanho variável mostram que oscilações coerentes, que aparecem apenas para estímulos que atingem um tamanho mínimo de cerca de 5° (dependendo do nível de contraste do estímulo). Estes resultados sugerem que oscilações rápidas da retina codificam mal mudanças sutis no tamanho do estímulo visual. Como nos estudos anteriores no geniculado lateral, registros obtidos diretamente da retina mostraram que oscilações rápidas da retina são altamente dependentes dos níveis de anestesia por halotano. E mais importante, em uma série de experimentos pode-se registrar respostas do geniculado lateral em um gato acordado, que foi subsequentemente anestesiado por halotano, mantendo-se o mesmo sítio de registro.

Oscilações rápidas da retina, ausentes durante a condição acordado, apareceram fortes como usualmente na condição de anestesia por halotano.

Estes resultados como um todo enfraquecem substancialmente a noção de serem as oscilações rápidas da retina importantes para o processamento visual. Por outro lado, demonstram que oscilações rápidas da retina podem apresentar propriedades semelhantes a oscilações gama no cortex. Desta forma, oscilações da retina induzidas por halotano podem servir como uma preparação interessante, mesmo se artificial, para o estudo da dinâmica de oscilações neuronais. !!

Palavras-chave: retina, geniculado, oscilação, coerência, halotano, acordado.

LI S T O F F I G U R E S A N D TA B L E S

Figure 1│ Seeing Matisse.

Figure 2│ Why Matisse would never build the collage to the right?

Figure 3│ Fast retinal oscillations.

Figure 4│ Maintained oscillatory responses in the retina and the LGN.

Figure 5│ Synchronization of oscillatory responses in the retina depend on size and continuity of

the stimulus.

Figure 6│ Oscillatory responses vanish in absence of halothane.

Figure 7│ An alert cat during a recording session.

Figure 8│ Head fixation apparatus and recording device X-Y table.

Figure 9│ Schematic representation of the electrodes and guide tubes.

Figure 10│ Recording device

Figure 11│ Visual stimuli used in the experiments in awake cats.

Figure 12│ Single-cell responses in the LGN are often oscillatory.

Figure 13│ Synchronization of ON and OFF-cell responses.

Figure 14│ Fast retinal oscillations arise from population interactions.

Figure 15│ Stimulus size and luminance modulate synchronous oscillations in single-cell responses

of the retina

Figure 16│ Single-cell contribution to a population rhythm.

Figure 17│ ON- and the OFF-oscillations are independent.

Figure 18│ Retinal oscillations in LGN vanish in absence of halothane.

Figure 19│ Oscillations are absent during ketamine anesthesia.

Figure 20│ In the awake cat, fast retinal oscillations are absent.

Figure 21│ Recordings in the LGN of an alert cat and during halothane anesthesia.

Figure 22│ Entrainment of responses to the refresh of a CRT monitor display.

Table 1│ Table 1.

SUMMARY

!AB S T R A C T A N D K E Y W O R D S …………..…….…….………………… 05

RE S U M O E PA L AV R A S-C H AV E …………..…….…….……………… 06

LI S T O F F I G U R E S A N D TA B L E S ………….….….….………………… 08

SU M M A RY ……………………..…….….….….…………………… 10

IN T R O D U C T I O N ……………………..…….….….….……………… 11

1.1 MAT I S S E S C I S S O R S ……………………………………..……… 11

1.3 OS C I L L AT I O N S I N T H E V I S U A L S Y S T E M……..…….……………….. 16

1.4 GR O U N D Z E R O …………………………………………………. 19

2. OB J E C T I V E S ……………………..……..….….……………… 21

Speci f ic goals ……….……….……………….……….…………… 21

3. ME T H O D S ………………………………………………………… 23

3.1 EX P E R I M E N TA L S E S S I O N S…………..…..…..…..…..…..…..…..…. 23

3.1 .1 SU R G I C A L P R O C E D U R E S……………………………….. 24

3.1 .2 RE C O R D I N G S……………………………………….…. 26

3.1 .3 DATA A C Q U I S I T I O N…………………………………..… 29

3.2 . V I S U A L S T I M U L I ………………………………………………… 30

3.4 . DATA A N A LY S I S …………………………………………………… 31

4. RE S U LT S ………………………………………………………… 34

4.1 . S I N G L E-C E L L A N A LY S I S ………………………………………… 34

4.2 . OS C I L L AT I O N DY N A M I C S ……………….………………………. 39

4.3 . DE P E N D E N C I E S O N H A L O T H A N E ……….………………………. 41

4.4 . NO-H A L O T H A N E C O N D I T I O N …………………………………… 41

4.5 . RE C O R D I N G S I N T H E AWA K E C AT………….…… ………………… 42

4.5 . ST I M U L U S E N T R A I N M E N T………….…… …………..…………… 44

5. D I S C U S S I O N ………… …………………………………………… 45

6. CO N C L U S I O N ………………………………………………………… 48

7. RE F E R E N C E S ………………...……………………………………… 49

8. TA B L E S……………………...……………………………………… 56

9. F I N A N C I A L S U P P O RT……….………………………………………… 57

10. AC K N O W L E D G M E N T S……….……………………………………… 58

!!

!

Do fast ret inal osci l lat ions play a role in v is ion? by Giovanne Rosso "11

!!1. IN T R O D U C T I O N

1 .1 MAT I S S E S C I S S O R S

The Matisse: The Cut-Outs exhibition at Tate in 2014, London, was a big

success. Half a million visitors came to see the bold colorful shapes, deceptively

simple, but majestic in their composition and force. Matisse has been always

loved for the warm, intense colors of his paintings. It is his cut-outs, however,

that vibrantly show a dialogue between texture, colors and forms. By cutting

shapes from colored papers, Matisse forges new dimensions of visual expression.

At the same time he exposes the very process of seeing:

It is no longer the brush that slips and slides over the canvas, it is the scissors

that cut into the paper and into the color. […] The contour of the figure springs from

the discovery of the scissors that give it the movement of circulating life. This tool

doesn’t modulate, it doesn’t brush on, but it incises in, […] because the criteria of

observation will be different. Henri Matisse1

In his collages, simple pieces of paper unfold their colors and shapes into

new fresh contexts. Matisse viewed them as virtual worlds, inhabited by flowers,

leaves and birds. So, it is not surprising to find mural-sized compositions in his

late work (Figure 1). He defined cut-out as „painting with scissors“, and saw in it

a source of liberating creativity and joy.

If for Matisse, seeing was a feast for the eyes, for the physiologist it may

represent life enduring questions. How shapes are cut from scenes, colors bound

to surfaces and pieces bound into wholes?

Admittedly, these are hard and largely unresolved problems. Yet, the last

decades have seen important conceptual and experimental advancements.

Basically we are confronted with two sets of questions. The first one refers to

the encoding of features, such as texture, color and shapes. When seeing the

mural in Figure 1, are colors and shapes processed together? Are curves and

1 quoted by Jodi Hauptman, in Henri Matisse The Cut-Outs Art-book, Museum of Modern Art, New York, 2014.


lines represented in the same way? Do we have dedicated circuits for seeing

faces? In the past years, some of these questions could be addressed rather

directly. An invaluable approach has been the characterization of response

properties of single cells, for which the work of Hubel and Wiesel (1961) is a

paradigmatic example. By recording neuronal responses in the visual cortex of

the cat, they discovered that responses can be very selective to specific stimulus

features, such as the orientation and the direction of movement. These seminal

findings triggered a full-blown research program, leading to a detailed account of

how the visual system breaks and encodes bits of visual information (Barlow,

1972). There is today a broad consensus on the parallel and hierarchical

organization of the visual system (Felleman and Van Essen, 1991; Gattass et al.,

2005). This body of knowledge ultimately explains how selective responses to

basic features (such as orientation, color, texture) are combined into higher-

order representations, such as for example during the recognition of complex

shapes or faces (Mazer and Gallant, 2000; Orban et al., 2004; Yovel and Freiwald,

2013).

Figure 1. Seeing Matisse. The Cut-Outs exhibition, Tate Modern, London, April 2014. Photograph by Guy Bell. © REX.


Still, a closer examination of Matisse’s panel reveals a radically different set

of questions. Why do all elements in a flower have the same color? Why colors

do not spill over into neighboring regions? Why we see some shapes as figures,

and some others as ground? Why do we see flowers or faces, and not something

else? Why are shapes sometimes pleasing and intense? These questions are more

difficult, and until recently eluded our best efforts. Essentially they broach

fundamental problems about large-scale integration in the brain. For a long time

we missed a clear understanding of how perceptions, thoughts and emotions are

put together from highly distributed networks. One obstacle comes from the

very dynamical nature of the cognitive processes. Perceptual or emotional

conjunctions are not fixed. On the contrary, they are distinctly contextual and

dynamical. When Matisse says that „the criteria of observation will be different“

he points out to the contextual nature of seeing (Figure 2).

Figure 2. Why Matisse would never build the collage to the right? Although the two pictures share the same local elements, globally they appear radically different. Perception is highly sensitive to context. Blue Nude IV, Henri Matisse, Spring 1952. Gouache on paper, cut and pasted, and charcoal on white paper 102.9 x 76.8 cm © Estate of H. Matisse 2014 (Tate Shop reproduction, London).

!


From perceptual grouping to contextual inference, shape recognition and

learning — soon it became clear that many aspects of cognition go beyond the

single-cell level. Put simply, one single electrode in the brain would not be enough

for understanding large-scale integration. Meanwhile, multi-electrode recording

techniques, championed by a handful of groups in the 60’s (Gerstein and Clark,

1964; Gerstein and Perkel, 1969, Freeman, 1975), became progressively a routine

in many laboratories worldwide. Today, even a modest laboratory has a high-

density, 128-channel neural signal acquisition system. Surely, this revolution was

only possible because of the accessibility of computers, everyday cheeper and

faster. Considerable efforts have also been made in the design of increasingly

large arrays of electrodes and the development of new bio-compatible implant

technologies.

The problem, however, was not only technological. Key concepts were

missing. What is the nature of the interactions in the brain? What are the

mechanisms that coordinate multi-scale activity bridging different levels in the

neural systems?

An attractive idea was put forward by Christoph von der Malsburg in the

early 80's (Malsburg, 1981; von der Malsburg, 1994; Singer and Gray, 1995). Briefly,

his proposal was that neuronal interactions come about in the time domain, at a

very fine scale (milliseconds). This conjecture was supported by abundant

experimental evidence (Singer, 1999; Buzsáki, 2006). Neuronal responses often

exhibit a fine temporal structure, characterized by periodic fluctuations or

oscillations (Gray and Singer, 1989; Singer, 1993). These observations paved the

now common notion that not only the rate, but also the timing of the action

potentials matters for cognitive processing.

In the visual system, oscillations were known to be an integral component

of the responses in various structures, and at different hierarchical levels. It was

in the cortex, however, that the observation of rhythms turned to be central for

our understanding of mechanisms (Eckhorn et al., 1988; Gray et al., 1989). The

core hypothesis was that the brain provides precise time relationships to build

active conjunctions at the perceptual level. In Figure 2, for example, the same set

of elements are bound into different percepts. In accord to the synchronization

hypothesis, the two figures results from different conjunctions defined by the


rhythmic firing of the neuronal ensembles. Seeing different compositions implies

in different synchronization patterns, even when the activation levels (rates) are

the same (locally the figures are the same).

The binding by synchronization hypothesis received copious experimental

and theoretical support (Singer, 1993). A first important finding was that

synchronization appeared to be restricted to fast frequencies, from 35 to 85 Hz

— the gamma band (Eckhorn et al., 1988; Gray et al., 1989). Another critical

result in these early studies was contextual sensitivity. Synchronization appeared

only for single-contour objects (such as a single coherent bar, Gray et al., 1989)

and not for contours moving in conflicting directions (Engel et al., 1991), even if

locally the stimuli were nearly the same. These experimental findings were in

agreement with the idea that synchronous oscillations provide a flexible

mechanism for perceptual segmentation and binding. Other studies in cats and

monkeys extended these conclusions, mainly by testing how global properties of

the stimulus modulates synchronization (Engel et al., 1991; Kreiter and Singer,

1996). Castelo-Branco et al. (2000) used bi-stable stimuli (moving plaids) to

demonstrate that response synchronization depends on the transparency of

superimposed surfaces. These results were relevant in showing that

synchronization can flexibly control the segregation of surfaces or objects.

A number of other studies, however, raised serious concern on the

significance of gamma oscillations for feature binding (Merker, 2013). Many

studies, mostly in behaving monkeys, failed to demonstrate a clear correlation

between gamma responses and perception (Thiele and Stoner, 2003; Roelfsema et

al., 2004; Palanca and DeAngelis, 2005; Lima et al., 2009; Ray and Maunsell, 2010;

Burns et al., 2011; Xing et al., 2102). Moreover, based on an information theory

analysis, it has been argued that that gamma oscillations arise because of

unspecific excitation–inhibition interactions in the networks, mechanistically

irrelevant for computations in the brain (Ray and Maunsell, 2015).

Notwithstanding these objections, beyond feature binding, gamma

oscillations have been associated with a number of other cognitive operations,

such as sensorimotor integration, attention, temporal expectancy and memory

(Fries et al. 2001; Womelsdorf et al., 2007; Lima et al., 2000; Engel and Fries,

2016). Gamma synchronization has been related to feature encoding (Vinck et al.


2010; Womelsdorf et al., 2012), a mechanism that could be complementary to

rate (Biederlack et al., 2006). Gamma responses have also been linked to

attention (Engel et al., 2001; Fries et al., 2001, see review in Fries, 2009) and

control of information flow in the brain (Akam and Kullmann, 2012). From an

engineering point of view, phase-locking of periodic signals are ideal for building

flexible relationships in highly distributed parallel networks, and may work as a

very basic mechanism controlling the flow of information in the brain (Akam and

Kullmann, 2014). This may explain why neuronal oscillations are ubiquitous across

diverse neural systems and well conserved during the evolution (Buzsáki et al.,

2013).

1.2 OS C I L L AT I O N S I N T H E V I S U A L S Y S T E M

High-amplitude rhythmic responses have been observed in the visual

system of different species as early as the beginning of the century (Gotch, 1903;

Einthoven and Jolly, 1908; Frohlich, 1914; Granit and Therman, 1935). Oscillations

have been found in retinal activity of various species of different vertebrates

groups, such as the frog (Ishikane et al., 1999), salamander (Wachtmeister &

Dowling, 1978), rabbit (Ariel et al., 1983), cat (Doty and Kimura, 1963; Laufer and

Verzeano, 1967 ; Arnet t , 1975 ; Neuenschwander and S inger, 1996 ;

Neuenschwander et al., 1999) and monkey (Doty and Kimura, 1963). Oscillatory

responses was found at different stages of the visual processing, from the retina

to the cortex. In the retina and the LGN, oscillations were observed in response

to large light stimuli (Neuenschwander et al., 1996), to an homogeneous

illumination of the whole visual field (Laufer and Verzeano, 1967) and

spontaneously, in maintained responses to light and in the dark (Bishop et al.,

1964; Laufer and Verzeano, 1967; Arnett, 1975; Neuenschwander et al., 1999).

Figures 3 and 4 shows examples of these early observations in the cat.

These early studies studies showed that oscillatory responses in the retina

are highly dependent on the size and the contrast of the stimulus. In the cat,

synchronization of oscillatory responses has been observed for distances large as

20 degrees of visual angle across the retina (Neuenschwander and Singer, 1996)

and were found in responses to all functional types (ON and OFF-cells, X- and Y-


cells) both in the retina and in the LGN (Ariel et al., 1983; Neuenschwander et

al., 1999; Ito et al., 2010).

In the cortex, gamma oscillatory responses (frequencies higher than

30Hz), were first reported in the 30s (Jasper and Andrews, 1938) as early as the

EEG recordings became available and by Adrian, in a study of the olfactory bulb

of hedgehogs (Adrian, 1942). These small variations in the EEG were originally

considered noise, especially when compared with the larger, slower rhythms and

evoked responses. More recently, however, there was a growing interest in fast

rhythms, which may play an important mechanistically role in perception and

cognition (Singer 2001; Singer, 1995; von der Malsburg, 1981).

It is important, however, not to mistake gamma oscillations with the retinal

oscillations. The term fast retinal oscillations are well used for those oscillations

in the 30-120 band generated in the retina but observed both in the retina and

the LGN recordings, since the LGN spiking patterns are inherited from retinal

ganglion cells activity (Sincich et al., 2007). Although fast retinal oscillations are

transmitted to the cortex, they do not necessarily contribute to generate

cortical gamma. In the study of Castelo-Branco et al. (1998) in the cat, data

obtained in simultaneous recordings from the retina, LGN and the cortex (areas

A17 and A18) show that gamma oscillations in the cortex follow a different

dynamics over time. Another important difference is that, in the cortex, gamma

responses are very sensitive to the orientation selectivity of the cells (Gray and

Singer, 1989), a feature that is not encoded in the retina.

Despite numerous studies it is still unknown, how the different features of

a visual scene (as in the Matisse’s cut-outs) are linked or segregated. Several

groups proposes that a possible and efficient mechanism for the linking of regions

and attributes that define the pattern could be based on temporal correlations,

and that the partial coherence of action potentials within a neural population

could be an operating principle for visual binding. Meanwhile, most of the studies

of fast retinal oscillations were made in the anesthetized and paralyzed cat. These

limitations raise questions about the role of oscillations in the retina. Do they

work as a binding mechanism or they are just epiphenomena? Are they an artifact

from the anesthesia?

!


200 ms

RETINA

RETINA

LGN

Figure 4. Maintained oscillatory responses in the retina and the LGN. Recordings made by Laufer and Verzeano in the 60’s (Laufer and Verzeano, 1967). Upper trace, mass activity recordings from the optic tract. Lower traces, simultaneous micro-electrode recordings from the retina (channels 1, 2 and 3) and the LGN (channel 4). Distances from the electrodes in the retina were about 200 µm. Ganzfeld illumination at 75 lux. Recordings were made in a non-anesthetized, paralyzed cat.

Figure 3. Fast retinal oscillations. Upper left, schematic representation of the experimental setup, recording system and correlation analysis implemented by D. Arnett in 1975. Synchronization was evaluated in realtime with a crosscorrelator device, which computed and displayed a crosscorrelogram between spike trains of two simultaneously recorded cells. Spikes were sorted from MUA signals by a logic circuit (spike classifier) based on time-amplitude window discrimination. Right, cross-correlograms computed for responses of a pair of ON-cells recorded in the LGN of a cat under halothane anesthesia (modified from Arnett, 1975; Figure 9; the bottom correlogram has a higher resolution). Lower left, mass activity traces from the optic tract in a non-anesthetized cat in response to steady illumination (Laufer and Verzeano, 1967; Figure 1).

SCOPE

TAPEREC

AUDIO PRE-AMP

SPIKECLASS

BOARD

MIRROR

CRT

PLOT

VIDEOGEN

TVMONITOR

X-CORRE

Time (ms)

32 HZ

32 HZ

0

0

-50 50

-500 500

A R N E T T, 1 9 7 5

L A U F E R A N D V E R Z E A N O , 1 9 6 7


1.3 GR O U N D Z E R O

In this study we focus on the role of fast retinal oscillations in visual

processing, without the constrains of anesthesia and paralysis. In the past years,

only a few experiments were made in the LGN to study oscillatory activity (Ito

et al., 2010), and we hoped to offer a fresh view on the results obtained in the

60’s in the non-anesthetized, yet paralyzed cat (Laufer and Verzeano, 1967).

Another motivation for studying the awake cat, however, came from an

observation we made recently in our laboratory, which was quite unexpected.

During an experiment in the anesthetized cat, we were forced to acutely

discontinue the halothane and replace it by ketamine (given i.m.). This happened

unwillingly, because of a failure in exchanging a gas bottle. The experiment was

running well, and as in many other occasions we were able to observe stunning

fast oscillations in the LGN.

A few minutes after withdrawing the halothane, however, for our surprise

and bewilderment, the oscillations vanished almost completely from the

responses (Figure 6). These unexpected findings prompt us to carry out a series

of new experiments to verify whether oscillation strength correlates with

Time (ms)

3° 3°

0-80 80Time (ms)

0-80 80

1

2

1

2

109 Hz 104 Hz

111 Hz

104 Hz

106 Hz

Figure 5. Synchronization of oscillatory responses in the retina depend on size and continuity of the stimulus. Notice that synchronization appears only the stimulus was continuous bridging the two receptive fields. RF distance, 6°. Responses from a pair of two ON-cells. Recordings were made directly from the retina, during halothane anesthesia (modified from Neuenschwander and Singer, 1996).


halothane concentration levels. We made also tests for isoflurane, which has

similar pharmacological properties. Our preliminary results in the LGN were

very conclusive. Retinal oscillations appears to be highly dependent on halothane

anesthesia.

Are fast retinal oscillations (at least in the cat) an artifact from the

halothane anesthesia? Do they play a role on vision as we initially thought

(Neuenschwander and Singer, 1996)? With the present study we want to provide

a definitive answer to these questions. Our main approach will be to characterize

the oscillatory behavior of the responses in the awake cat, under naturalistic

conditions, such as during free-viewing of a visual scene, and compare to data

obtained during anesthesia by halothane (or isoflurane) and by ketamine (control

experiments).

Figure 6. Oscillatory responses vanish in absence of halothane. The horizontal stripes in the sliding window analysis plots (left) reveal the oscillatory modulation of the responses, which very strong for a halothane concentration of 1.0%. Notice that removing the halothane leads to a slight increase in response levels (histograms, right). Oscillation frequency, 72 Hz. The stimulus was a bright disc presented over the RFs, with luminance increasing linearly. Stimulus size, 12°(from Freitag, 2013; Figure 20).

12°

cgl04c0502 5-580

-80

80

0.40

0.0

0

-80C

oin

cid

ence

s

T ime3 0 0 m s

cgl04c0503 5-5

Tim

e (m

s)

0

cgl04c0501 5-580

0T ime (s )

1.0 3.0

100

cgl04c0502 5-5

cgl04c0503 5-5

cgl04c0501 5-5

1.0%H A L O T H A N E



2.0

Sp

ikes

/s


2. OB J E C T I V E S

!The primary goal of this study was to test whether fast retinal oscillations

are present or not in conditions compatible with natural seeing. This involved (1)

recording directly from the retina of the anesthetized cat to study spiking

responses to dynamical stimuli, such as natural scenes movies and varying-size

stimuli and (2) obtain data from the alert cat, removed from any influences of

anesthetic agents.

Recently we have shown that halothane and isoflurane are responsible for

the generation of fast rhythms in the retina. This first study, however, was limited

to an autocorrelation analysis of multiunit responses in the LGN. Here, we aim to

extend and refine these results by recording single-cell activity simultaneously

from the retina and the LGN. Comparisons will be made for recordings under

halothane (isoflurane) and ketamine anesthesia, and also for recordings under

ketamine anesthesia without previous exposition to halothane (ketamine-only

condition). Finally, these results will be compared to data obtained in the alert

cat.

Quantification of the oscillatory dynamics will be based on a sliding-

window correlation analysis and multiaper spectrum and coherence of single-cell

spiking responses.

!Speci f ic goals

!• To compare the oscillatory behavior of the retinal responses under

halothane (or isoflurane) and ketamine anesthesia;

• To verify at the single-cell level whether breaking of stimulus continuity

disrupts synchronous oscillatory responses;

• To verify at the single-cell whether synchronous oscillatory responses are

also present in responses when probed with dynamical stimuli, such as

natural scene movies and size-varying stimuli.


• To record LGN responses in the cat under ketamine anesthesia without

previous exposition to another anesthetic (ketamine-only condition).

• To record LGN responses in the awake cat.

!!!!!!!!!!!!!!!!!!!!


3. ME T H O D S

Adult cats from our colony (BSIC-Instituto do Cérebro - UFRN) were

used in this study (N= 5).

All experimental procedures were approved by the ethical committee for

animal experimentation of the Universidade Federal do Rio Grande do Norte

(CEUA-UFRN protocol nº 019/2012) and were in compliance with the guidelines

of the European Community for the care and use of laboratory animals

(European Union directive 86/ 609/EEC).

3.1 EX P E R I M E N TA L S E S S I O N S

In a first set of acute experiments, recordings were made from the LGN

and the retina of anesthetized and paralyzed cats (N= 3). Different anesthetics

were used during the same experimental session. Generally we started a session

with ketamine (induction), followed by halothane, which subsequently could be

replaced by isoflurane or combined with ketamine. Ketamine was always applied

if halothane or isoflurane were to be absent (usually less than 1 hour periods),

assuring a surgical plane of anesthesia during all procedures. In these experiments

the cats were not recovered at the end of the recordings, which typically lasted

for 96 hours (4 continuous days).

In addition to this group we were able to record from the LGN of cats

(N= 2) without any immediate exposition to halothane (or isoflurane). For this, a

recording chamber was chronically implanted on the skull. Typically, these cats

were submitted to multiple recording sessions (~ 10 sessions), which lasted for 3

to 4 hours. In one series of experiments, recordings started after ketamine

anesthesia, the ketamine-only condition (since previously the cat received no

other anesthetic than ketamine). In another series, recordings were made in the

awake cat (N= 1), without influence of any anesthetic agent, the awake condition.

To this aim, one cat was habituated over several months to sit quietly with its

head fixed for 2 to 3 hours (Figure 10). During the training and recording

sessions cat food rewards were always given abundantly. In 2 occasions, we were

able to sample data from the same LGN recording sites for all three conditions

(awake, ketamine-only, halothane).


Figure 7. An alert cat during a recording session. In our study one cat (shk) was trained to sit quietly while having the head fixed. We were able to obtain stable recordings from the LGN over a few hours. Notice that the cat was fixed by the two recording chambers. In general cat with their heads fixed stare at the monitor, making possible to coarsely map the RFs (see example in Figure XX).

Figure 8. Head fixation apparatus and recording device X-Y table. The head of the cat was held by two identical chambers chronically implanted on the skull (only the recording chamber is shown). The fixation system had a long column (see Figure 10), in which a L-shape plate was mounted (model shown to the left). A bored cylindrical adapter was screwed to the recording chamber (14 mm in diameter, 1.0 mm thread). The X-Y table and recording device were than attached to the L-plate with this adapter (see Figure 12). 3D modeling by Heitor Bernardino de Oliveira, Instituto do Cérebro - UFRN, Natal.

X-Y Table

Device guiding bars

Guide tube positioning scale

Frame column

L-shape headholder

Adaptor

AP-positioning knob

Recordingchamber


3.1 .1 SU R G I C A L P R O C E D U R E S

For the acute recordings a single large recording chamber (15 mm in

diameter) was surgically cemented on the skull at the beginning of the

experiment. Generally, this procedure took a few hours to be completed. After

that, the cranium was opened, the dura removed and the cortex above the LGN

exposed. Electrodes were than positioned into the LGN (centered on Horsley-

Clarke coordinates AP 6 to 7, ML 9 to 10), and the recording sessions started.

Before the surgical procedures, the cats received atropine sulphate

(Atropion, Ariston, Brazil, 0.1 mg/kg i.m.) and were sedated with Xylazine

(Xylazin, Syntec, Brasil, 0.25 mg/kg i.m.) combined with Ketamine (Cetamin,

Syntec, Brasil, 10 mg/kg, i.m.). The cats were than intubated (Braun cuffed

endotracheal tube, Germany, 3.5 to 4.0 mm) and artificially ventilated. Anesthesia

was maintained with 0.8 to 1.2% halothane (Halotano, Hoechst do Brasil) in a

mixture of oxygen (30% to 40%) and nitrous oxide. The volume of the respiration

pump (35 to 45 ml) and the respiration frequency (14 to 20 stokes/ min) were

adjusted to yield a ventilation pressure of 7-10 mbar and expiratory CO2 in the

range of 2.6 to 3.5%. A rectal thermometer connected to a heating pad unit was

used to maintain body temperature at 38°C. Relevant parameters for life support

(EKG, temperature, expiratory CO2 and SpO2 trends, inspiratory and expiratory

halothane and oxygen concentrations) were monitored continuously by means of

a patient monitor (Dash 3000, linked to a Smart anesthesia multi-gas unit, GE

Heathcare, USA). Fluid loss was compensated by infusion of saline solution

(Braun infusion pump, Germany, 6 ml/h i.v.).

In order to record from the retinal ganglion cells we employed the

intraocular recording technique originally developed by B. G. Cleand in Camberra

during his seminal work in the retina (Cleland et al.,1971), and posteriorly

modified by Heinz Wässle in Konstanz (Wässle and Peichel, 1979). We used a

modified stereotaxic frame (Wässle, 1975), which left the orbit of the cat free,

thus, making easy the placement of the recording device. After opening the skin

laterally to the canthus, the conjunctiva around the eye ball was cut and the

sclera exposed. A steel ring was then fixated to the sclera just behind the limbus

by means of 5 to 7 modified Donati stitches. The stitches were distributed

equally around the globe, assuring a strong bound between the sclera and the


ring, which was hold in place by a long articulated horizontal arm connected to

the stereotaxic apparatus. By releasing a single screw, one could rotate the ring

to a desired position, and consequently the eye. We used a fundus camera (Zeiss,

Germany) to determine the best position of the eye as a function of the retinal

landmarks (the area centralis and the optic disc) which were projected on the

computer screen used for visual stimulation. A bored plate fixed to the ring

supported the recording device, so no pressure was applied to the eye. After

opening the sclera with a cauterizer (Fine Science Tools, Germany), the

electrodes (mounted in individual guide tubes) were inserted into the eye though

a cannula (1.2 x 10 mm). The apparatus had a spherical bearing allowing angular

rotations of the cannula in the posterior chamber of the eye. In this way, with

help of the fundus camera, we could aim the electrodes to almost any desired

Figure 9. Schematic representation of the electrodes and guide tubes. The quartz electrode is connected to the glass piston and the cable with a soldering pellet (melted by a hot air blower). A staple can be used for the L-shape arm element. Notice that the electrodes can be loaded into the guide tubes from above, simplifying exchanges. Electrode length, 100 mm. Distance between the grid and guide tube, 45 mm. Glass piston, 30 mm, diameter of 2.7 mm. For clarity elements are not drawn with the same scale. Developed jointly by Bruss Lima, Sergio Neuenschwander, Jerome Baron and Johanna Klon-Lipok at the Max-Planck Institute for Brain Research, Frankfurt.

Piston (glass)

Soldering pellet

Fixation ring

Narishige microdrive

Microdrivepiston

Needle 0.6 mm(steel)

Quartz electrode

Needle 0.3 mm (steel) 2X

Heat shrinkingtube

Grid (nylon)

Plate (Plexyglas)

Pin connector

Cable

L-shape arm(steel)


location in the retina. This technique yielded very stable recordings from the

retinal ganglion cells. Experiments were discontinued only when the optics of the

eye start to deteriorate, what usually happened in 2 to 3 days.

For the semi-chronical recordings two small titanium chambers (6 mm in

diameter) were surgically implanted on the skull at the position of the LGN in

the two hemispheres, respectively. The chambers were identical, but only one of

the two was actually used for the recordings. After opening the skin and exposing

the cranial vault, 10 to 13 self-tapping titanium screws (Synthes standard cortex

screws, Germany, 2.0 mm) were placed into the bone following a horizontal plane

just above the zygomatic arc. Acrylic cement (Paladur, Heraeus Kulzer, Germany)

was then spread in successive layers, to build a prothesis anchoring the chambers

and fixation screws. We observed a one-month recovery period before starting

the recording sessions. Typically recording session were scheduled one every 1-2

weeks. Implanted cats were not hindered by the prothesis. Local infections were

controlled by daily cleaning the skin borders with saline and topic oxygenated

water. These cats led a normal life among the others in the colony, with no signs

of discomfort or distress (one of our cats is already implanted for more than 1

and 1/2 year).

3.1 .2 RE C O R D I N G S

This study is entirely based on extra-cellular recordings of action

potentials (multi-unit and single cell activity).

We used quartz-electrodes (tungsten-platinum fiber electrodes insulated

by quartz, Thomas Recording, Germany, 80 μm in diameter). These electrodes are

known to have a good signal to noise ratio, and are rigid enough to penetrate the

brain (or the vitreous) undeviatingly.

In all experiments we employed a customized recording device (designed

by Sergio Neuenschwander at the Max-Planck Institute for Brain Research,

Frankfurt). Essentially it consists of 5 oil hydraulic microdrives (MO-95, Narishige,

Japan) mounted in a movable platform (Figure 12). The quartz electrodes are

placed in single guide tubes, which are mounted into a grid (determines the

spacing of the electrodes). A glass capillary mounted at the end of the guide

tubes serves as a piston for moving the electrodes. These pistons are connected


to their respective drivers by teflon elements. Our recording device allows for

the placement of the guide tubes to a desired depth (all together, with their

electrodes). The electrodes, in turn, can be moved independently (fine

displacements controlled by the microdrive units) or as a group (coarse

displacements controlled by a vertical screw). As shown in Figure 13 the guide

tubes consists of two thin needles (Ehrhardt Supra, Germany, 0.3 x 23 mm)

mounted inside a thicker cannula (Braun 100 Sterican, Germany, 0.6 x 60 mm)

and glued in place with an instant adhesive (Super Bonder, Loctite, Brasil). The

thin needle provides a a good cutting edge when moving the guide tubes through

the tissue.

The recording device was fit with a X-Y positioning system (part of the

Narishige MO-95 recording system), allowing for systematic positioning of the

electrodes in the horizontal plane (see Figure 13). This was particularly useful for

localizing the LGN. Our recordings were aimed at the region of central

representation of the visual field (less than 10 degrees of eccentricity).

For the recordings from the LGN, the guide tubes were first placed 5 to 7

mm above the the nucleus and than moved slowly, one by one, until lamina A was

Figure 10. Recording device. Our device allowed for independent positioning of the guide tubes and the electrodes. It was mounted on a X-Y table which was attached to the head-holder plate by a cylindrical adapter (See Figure 11). Device structure was made in PEEK. Designed by Sergio Neuenschwander at the Max-Planck Institute for Brain Research, Frankfurt. 3D modeling by Heitor Bernardino de Oliveira, Instituto do Cérebro - UFRN, Natal

Microdrive

Electrode coarsepositoning scale

Electrode coarsepositoning knob

Fixation ring

Grid

Glass piston

Guide tube

AP-Guide tube positioning knob

Pexyglas plate

Stopper

L-shape arm


found (robust reposes to the contralateral eye). Typically 3 to 4 electrodes were

used for the LGN recordings, both in the anesthetized and alert cats.

For the intraocular recordings we used a similar approach. The guide tubes

penetrated the vitreous in the posterior camera, and were placed a few

millimeters above the retina. Than they were moved individually under visual and

acoustic control (listening to the recorded spiking activity), until they reached

the retinal ganglion cell layer. Usually 2 to 3 electrodes were used for the

recordings in the retina.

In all experiments in which inhalation anesthesia was applied (either acute

or semi-chronical) eye movements were blocked by the intravenous infusion of a

paralyzing agent (pancuronium bromide, Nova Farma, Brazil, loading dose of 0.5

mg/kg i.v., maintenance dose of 0.25 mg/kg/h i.v.). After paralysis, the pupils were

dilated with topical application of atropine sulfate (Atropine-POS, Ursapharm,

Germany, 1%) and the nictitating membrane retracted (Neosynephrin, Ursapharm,

Germany, 5%). The cornea was protected with contact lenses containing artificial

pupils of 2 mm diameter. The eyes were focused on the stimulus monitor with

add of correcting lenses whenever necessary (Rodenstock manual refractometer,

Germany).

3.1 .3 DATA A C Q U I S I T I O N

Spiking activity from neuronal groups (MUA) were recorded after

amplification and band-pass filtering of the compound signals (0.7 – 6.0 KHz)

with a 32-channel Plexon modified preamplifier and a HST16025 headset (Plexon

Inc, Dallas, TX, USA). Data acquisition made with the SPASS software (written in

LabVIEW by Sergio Neuenschwander at the Max Planck Institute for Brain

Research), based on M-series NI acquisition boards (National Instruments, USA).

Signals were sampled at 32 kS/s with an additional 10 X onboard amplification.

Spikes were detected after a simple amplitude threshold algorithm, typically set

to twice the noise level.

The SPASS software provides modules for on-line visualization of the spike

waveforms, response trends and on-going autocorrelation function. Because our

analysis is focused at the single-cell level, we generally adjusted the position of

the electrodes to yield maximal responses and big spike waveforms. The online


information of the temporal structure of the ongoing responses was also very

useful to guide the experiments.

!!!!3.2 . V I S U A L S T I M U L I

Visual stimuli were presented on a 21” CRT monitor (Hitachi, CM803ET),

placed about 57 cm from the cat. Refresh rate was set to 100 Hz (except for the

protocols used for evaluation of stimulus entrainment, see Results section) at a

resolution of 1024 x 768 pixels (1.0° of visual field corresponded to 25 pixels).

S t imulus presentat ion was contro l led by the Act iveSt im sof tware

(www.activestim.com). Protocols consisted of a series of 10 to 250 repetitions

(according to the number of conditions of each stimulus protocol). For all

protocols, the different stimulus conditions were presented in a random order.

At the beginning of the recording sessions, RFs were searched with a

variety of stimuli, such as black and white cardboard, the experimenter’s hands

(Figure 7) and a handheld DC-light projector. If robust responses were found we

proceeded to an automatic mapping of the RFs (Fiorani et al., 2014; see a

comparison of various mapping methods in Pipa et al., 2012). Essentially this

procedure consisted in presenting a high-contrast bar (10 X 1000 pixels) at 16

different directions of movement (step of 22.5°). RF maps were obtained by

computing a response matrix with 10 ms resolution, corresponding to

approximately 0.2° in visual angle (see example of Figure XX). Generally, stimuli

LIGHT PATCHGRAY-LEVEL BINARY

Figure 11. Visual stimuli used in the experiments in awake cats. Stimuli were large covering circa of 20° of visual angle centered at the computer screen. Gray-level natural scene movies had 200 luminance values. Binary natural scene movies were displayed with 2 luminance values (black and white). All stimuli were smoothed spatially to avoid responses to high contrast borders. Circle indicates typical RF position relative to the stimulus.

http://www.activestim.com


were centered on the receptive fields altogether (only in selected cases the

stimuli were centered on individual receptive fields).

3.4 . DATA A N A LY S I S

Our analysis is focused on evaluating the oscillatory behavior of single-

cells responses (SUA), which could be grouped to yield a small population signal

of same response polarity (ON or OFF-cells MUA). In our analysis we used the

NEUROSYNC package developed in LabVIEW (National Instruments, USA) by

Sergio Neuenschwander and Jerome Baron. Additionally, we used Matlab

(MathWorks, USA) routines of the Chronux (www.chronux.org), an open-source

analysis software (see discussion in Mitra, 2007), which were embedded in the

LabVIEW environment. Spike sorting was carried out with SpikeOne (a LabVIEW

program written by Sergio Neuenschwander) relying on principal component

analysis of the spike waveforms and k-means clustering analysis (Machine

Learning LabVIEW toolkit). Numerous visualization and analytical tools, such as

the refractory period seen in autocorrelograms, were available to further guide

the refinement of the sorting (merging of clusters, exclusion of spike waveforms).

The oscillatory behavior of the responses was first assessed in the time

domain. For all data we carried out an average sliding window correlation analysis

(200 ms window in 50 ms steps), so we could follow the oscillatory behavior of

single-cells over time. This analysis proved to be very useful for the sorting

refinement, since the ON and OFF components could be easily identified in the

responses (see example in Figure 14). Trends and discontinuities in the oscillation

strength, frequency, phase were quantified by computing average auto- and cross-

correlations of SUA and MUA within 500 ms windows (sometimes we used

shorter windows as indicated in the figures).

Quantification of oscillation strength and frequency were made in the

spectral domain. We use the multitaper Chronux functions mtspectrumpb and

coherncypb for spectral analysis and coherence, respectively.

In brief, multitapering methods attempt to reduce the variance of spectral

estimates by multiplying the data with several orthogonal tapers (slepian

functions). Therefore, the frequency decomposition of the data yields

independent spectral estimates which is less sensitive to noise. The multitapered

http://www.chronux.org


power spectrum of a time series is defined for a given frequency as an average

across all repetitions and tapers:

!!!!

where

!!

is the discrete Fourier transform of the product of the measured time series

sequence { �� , n = 1, 2, ..., �� } with the �� -th slepian taper, denoted by

�� . Numerically, �� is computed as the FFT of the product. In our

analysis data were padded with zeros to the length of 2048 before the Fourier

transform. Five slepian tapers were used. Thus, we obtained a spectral resolution

of ±5 Hz and ±15 Hz for a 500 ms and a 200 ms window, respectively.

Synchronization of the oscillatory responses was evaluate by the

coherence, defined as:

where �� and �� are the multitapered power spectrum estimates of the

time series �� and �� averaged over n repetitions, respectively, and ��

is the cross-power of these two time series. Coherence provides a normalized

metric of linear dependencies between two processes, scaling from 0.0 to 1.0.

For a noiseless data, a coherence value of 1.0 should be obtained at all

frequencies if two processes are linearly related (i.e., their amplitude covary and

Sx f( ) Sy f( )

xn t( ) yn t( ) Syx f( )

Cyx f( ) =Syx f( )

Sx f( )Sy f( )


they share a constant phase relationship). If the two processes are completely

independent, coherence should be equal to 0.0.

The 95% confidence bounds for the spectral estimates were determined

by the jack-knife method across tapers and trials, as implemented in the Chronux

software.


4. RE S U LT S

In a first set of experiments, carried out in cats anesthetized with

halothane, we describe at the single-cell level the characteristics of oscillatory

responses in the LGN. We employed dynamical stimuli, such as varying-size bright

discs and movies, to follow the synchronization behavior of the responses over

time (e.g., oscillation strength and frequency). Our single cell analysis allowed us

to follow independently the ON and OFF-components along the responses.

In a second set of data, we present the effects of varying the

concentration level of halothane. Occasionally we also employed isoflurane, an

halogenated anesthetic similar to halothane. As a control, we compare the effects

of ketamine, either combined to halothane (or isoflurane) or after the halothane

withdrawing test.

By recording directly from the retinal ganglion cells with intraocular

electrodes, we show direct evidence whether halothane affects the generation of

oscillations within the retina (and not at the thalamic level). In a few cases

recordings were carried out simultaneously with the LGN, enabling us to follow

the effects of anesthesia a the two levels, retina and thalamus.

In a last series of experiments, data were obtained in absence of

halothane. In two cats, recordings were made from the LGN under ketamine

anesthesia, without previous exposition to halothane (the ketamine-only

condition). Finally, we present LGN data obtained in a freely viewing cat.

4.1 . S I N G L E-C E L L A N A LY S I S

Single-cell responses in the LGN are often oscillatory. In Figure 12,

recordings were made from Lamina A of an anesthetized cat with halothane. A

sliding window correlation analysis reveals very strong oscillations for cell

responses of both ON and OFF-polarities. The light stimulus evoked strong

responses in one ON-cell (1a), which persisted for a few hundred milliseconds.

Notice, however, that oscillations are absent at the very transient component of

the responses to the onset of the stimulus (seen as a sharp peak in the response

traces). This was characteristic in most of the recordings (see examples in

Figures 13, 14 and 15). Likely, oscillations may take up to 100 msec to build,


persisting afterwards in the late components of the responses. Interestingly, the

responses of the OFF-cells to the offset of the light stimulus appeared

instantaneously, without the characteristic non-oscillatory component of the

ON-cells (Figure 12, unit 1b; Figure 3, units 1b-2a in the retina). These

differences were consistent among many of our recordings, despite the

considerable smoothing inherent to our sliding analysis, which may hide sharp

transitions along the responses (typically we used a widow of 200 msec with a

step of 20 msec). Contrary to previous observations from MUA responses

(Neuenschwander et al., 1999), our single-cell analysis revealed no significant

differences in oscillation frequencies for the ON- and OFF-cell responses (see

examples in Figures 12 and 13), in accord with the findings of Ito et al., 2010).

Likely, fast oscillations in the retina arises from interactions among

neighboring retinal ganglion cells. A strong evidence is shown in Figure 14.

Recordings were made in the retina under halothane anesthesia. Auto and cross-

0.5

0.0

0.5

0.0

Tim

e (m

s)

0

-80

80 cgl03e12 5a-5a

0

-80

80 cgl03e12 5b-5b

0.3

0.0

Co

inci

den

ces

spik

es/s

T ime5 0 0 m s

0

-80

80 cgl03e12 5g-5g

5 0 0 m s

OFF-cell

1b

ON-cell1c

10

1a

ON-cell

11°

Figure 12. Single-cell responses in the LGN are often oscillatory. Strong oscillatory responses were observed for single-cells of both ON- and OFF-polarity (upper and middle sliding correlation analysis plots). Notice that not every cell oscillates. The response of the ON-cell shown in the lower plot, although robust and sustained, has no signs of temporal structure. Response traces are shown to the right. Responses to a bright disc flashed over the RFs. Stimulus size, 11°. Anesthesia, halothane.


Tim

e (m

s)

0

80

-80

nal004l04 17a, 17b, 18a, 18b

Coi

ncid

ence

s

0.2

0.0

74 Hz1.9 ms

0-80 80 804010 120

804010 120

Tim

e (m

s)

0

80

-80

nal004l04 17a, 17b, 18a, 18b

Coi

ncid

ence

s

0.2

0.0

78 Hz-0.8 ms

0-800.0

0.3

0.0

0.3

0.0

1.0

0.0

1.0

80Time (ms) Frequency (Hz)Time

1a - 2bR E T I N A

ON-cells

1b - 2aR E T I N A

OFF-cells

Figure 13. Synchronization of ON and OFF-cell responses. Simultaneous recordings from 2 pairs of cells in the retina. Each pair of units were obtained from separated channels. Synchronization of the ON-cells are shown above, while the OFF-cells below. Left panels, average sliding window analysis. Middle panels, average cross-correlation function computed within a 500 ms window (indicated by the black bar in the sliding correlation plot). Right panels, multitaper coherence analysis. ON-cell oscillation frequency, 74 Hz. OFF-cell oscillation frequency, 78 Hz. Responses to a bright disc flashed over the RFs. Stimulus size, ~20°. Anesthesia, halothane.

Tim

e (m

s)

0-50 50Time (ms)

50

- 50

0

Tim

e (m

s)

50

- 50

0

shk011j02 3a, 3b, 3c, 3d shk011j02 3a, 3b, 3c, 3d shk011j02 3a, 3b, 3c, 3d

Co

inci

den

ces

Co

inci

den

ces

Co

inci

den

ces

0.25

0.2

0.0

0.2

0.0

0.2

0.0

0.1

0.0

0.25

78 Hz

78 Hz

79 Hz

78 Hz

78 Hz

-0.2 ms

2.2 ms

1.2 ms

1.2 ms

1a - 1a

1a - 1b

1a - 1c

1a - 1d

1a - (1b,1c,1d)

1a

1b

1c

1d

(1b,1c,1d)

TimeTime

LGN

Figure 14. Fast retinal oscillations arise from population interactions. Data obtained from 4 cells recorded simultaneously from the same electrode. Left panels, sliding autocorrelation functions. Right panels, cross-correlation functions between unit 1a and each one of the 3 other units (1a-1b, 1a-1c and 1a-1d). Observe that for the individual neurons oscillatory patterns were often weak and discontinuous. The central bin in the crosscorrelogram is equal zero because superimposed spikes were discarded in the spike sorting process. Oscillation frequency, 78 Hz. Responses to a light circle. Stimulus size, ~20°. Anesthesia, halothane.


correlation analysis show that even though individual cells may exhibit weak or

discontinuous oscillatory patterns, at the population level, oscillations are strong

and stable (oscillation frequency, 78 Hz). This means that, despite firing in an

strong oscillatory manner, the individual cells may skip many cycles, a feature also

described in the cortex (Nikolic, 2013).

However, it needs to be emphasized that not all cells of same polarity

contributed to the oscillations seen in the MUA signals. A clear example is shown

in Figure 12 (unit 1c). While one of the two ON-cells recorded exhibited a very

strong oscillation, the other showed no signs of oscillatory patterning, even

though the two response is equally robust. Interestingly these two cells have

clearly different response profiles (unit Ic exhibits a sustained response

compatible with a X-cel, while unit 1b exhibits a transient response, compatible

with Y-cell). Although beyond the scope of this study it would be interesting to

see whether the different functional types (X or Y-cells) are capable of

synchronizing their responses, depending on the characteristics of the stimulus.

nal004l01 17a 18b

Tim

e (m

s)

80

0

-80

Coi

ncid

ence

s

0.1

0.0

Time

0.9

2°

4°

6°

8°

10°

12°

0.6 0.3Luminance

1a - 2aSize

1°

nal004k02 18c

nal004k02 17a, 17c

1aON-cellRETINA

ON-cell2a

RETINA

1°

Figure 15. Stimulus size and luminance modulate synchronous oscillations in single-cell responses of the retina. Cross-correlation sliding window analysis show a strong correlation between the size and the stimulus and oscillation strength. Data obtained for 3 different luminance levels (175, 80, 50 Lux).


As previously described for MUA responses (Neuenschwander et al.,

1999), synchronous oscillations in the retina and the LGN were very sensitive to

these size of the stimulus. In Figure 15, a strong correlation between stimulus

size and oscillation strength. We used 6 different sizes of the stimulus at 3

luminance levels (175, 80 and 50 Lux, corresponding to contrast ratios of 0.9, 0.6

and 0.3). From the plots in Figure 15, it is obvious that oscillation strength

increased non linearly as a function of stimulus size. At a relatively high luminance

level (50 Lux in our experiments, contrast of 0.3) only a circle size greater than

6° degrees sufficed to trigger oscillatory responses. Likely, a critical size value had

to be reached for the spreading of oscillation among activated mass of retinal

ganglion cells.

In our experiments, synchronization was always accompanied by

oscillations. For a pair of cells Depending on the stimulus conditions, the

coherence values could be surprisingly high, near the maximal value of 1.0 (see

examples in Figure 13 and 16), indicating that spikes exhibited very consistent

phase relations.

!!!

40 120Time

80

120

40

TimeFrequency (Hz)

80

120

40

Freq

(Hz)

80

120

401a - 2b

R E T I N AON-cells

500 ms

0

80

-80

nal004l04 17a, 18b (2)

Coi

ncid

ence

s

0.2

0.0

Coi

ncid

ence

s0

80

-80

nal004l04 17a, 18b (4)

0.2

0.0

Tim

e (m

s)

0

80

-80

nal004l04 17a, 18b (3)

Coi

ncid

ence

s0.2

0.0

0.0

1.0

83 73 72

0.0

1.0

84 73 81

0.0

1.0

82 75 70

Join

t ra

te (

sp/s

)

20

Rat

e (s

p/s

)

20

20

Figure 16. Modulation of the oscillatory to a size-walk stimulus. The size-varying stimulus led to a very strong modulation of the responses. The coherence between the two cells was much less sensitive to the variation in size, not following the rates.


4.2 . OS C I L L AT I O N DY N A M I C S

To test how robust would be an oscillation-based encoding mechanism in

the retina, we followed the oscillation dynamics (strength, frequency and phase)

of responses to dynamical stimuli, such as size- or luminance-varying bright discs

(random walks) or natural scene movies. In the experiment shown in Figure 16,

we used 3 different size-varying functions for a bright disc (size-walk stimuli)

presented over the RFs of two ON-cells in the retina. As expected, the size-walk

stimulus led to a very strong modulation of the responses. Interestingly, the

coherence between the two cells was much less sensitive to the variation in size,

definitively not following the rates (see middle panels in Figure 16). Moreover,

the oscillation frequency tend to decrease smoothly after the onset of the

oscillations (starting around 200 msec after the appearance of the stimulus). This

decay in frequency was probably due to a single global oscillatory process,

because in general there was no discontinues or transients in frequency or phase

(see frequency plots in Figures 16). It has been observed in virtually in all data, in

the retina and LGN, and can be considered as a hallmark of the fast retinal

oscillations (see also examples in Figures 13 and 15).

Time

80

120

40

Time

80

120

40

Freq

(Hz)

80

120

40

1a

(1a, 1b)R E T I N A

ON-cell

1bOFF-cell

500 ms

0

80

-80

Coi

ncid

ence

s

0.2

0.0

Coi

ncid

ence

s0

80

-80

cgl07a03 17a (3)

cgl07a02 17a

cgl07a02 17ab

cgl07a03 17a, 17b (3)

cgl07a03 17b (3)

0.1

0.0

Tim

e (m

s)

0

80

-80

Coi

ncid

ence

s

0.2

0.0

Rat

e (s

p/s

)

50

Figure 17. ON- and the OFF-oscillations are independent. Cross-correlation sliding window analysis for two cells of opposite polarity recorded in the retina.


However, depending on the size history of the stimulus (and consequently

on the size history of the population of active cells), the oscillation process could

be reset. In Figure 16, when the stimulus decreases in size so extremely that the

cell responses cease (see arrowhead in the frequency plot), there is a jump in the

oscillation frequency for the upcoming response (from 73 to 81 Hz), at the very

moment the stimulus reaches de novo a critical size.

Resets in global ongoing oscillations were found both for ON and OFF-

cells. An intriguing example is shown in Figure 17. In this case the size-walk

stimulus was centered at a point outside the RFs of two cells recorded in the

retina. The cells had opposite polarity and overlapping RFs. Thus, when the size-

walk stimulus invaded the RFs the ON-cell fired while the OFF-cell silenced. The

inverse occurred for when the stimulus left the RFs. This explain why the

responses barely overlapped. There were a few resets in the oscillations for both

the ON and the OFF-responses. Remarkably, the oscillations resets were

independent of each other, indicating that the ON- and the OFF-oscillations do

not share a common input.

T imeT ime

1.0%

0.6%

0.2%0.3

0.0

cgl03e07 5 (2)

cgl03e06 5 (2)

cgl03e08 5 (2)

cgl03e07 5 (3)

cgl03e06 5 (3)

cgl03e08 5 (3)

0

-80

80

Co

inc

ide

nc

es

0

-80

80

0

-80

80

500 ms

Tim

e (m

s)

0.3

0.0

0

-80

80

Co

inc

ide

nc

es

0

-80

80

0

-80

80

500 ms

Tim

e (m

s)

20° 20°

HALOTHANE

Figure 18. Retinal oscillations in LGN vanish in absence of halothane. Left panels, sliding window correlation for responses to a natural scene movie. Right panels, responses to a large patch of light. Observe that the effects on oscillation strength do not correlate linearly with the concentration levels of halothane (already at 0.6% oscillations ceased almost entirely). Firing rates are slightly augmented after halothane withdraw, indicating that halothane causes a slight depression in the general activity of the retinogeniculate system.


4.3 . DE P E N D E N C I E S O N H A L O T H A N E

As described recently for MUA responses (Freitag, 2013), single-cell

oscillations are unambiguously correlated to the level of halothane anesthesia.

Figure 18 documents how varying the halothane concentration affected

the oscillatory behavior for SUA recordings in the LGN for responses to a natal

scene movie and a large patch of light. Observe that the effects on oscillation

strength do not correlate linearly with the concentration levels of halothane

(already at 0.6% oscillations ceased almost entirely). Firing rates are slightly

augmented after halothane withdraw, indicating that halothane causes a slight

depression in the general activity of the retinogeniculate system. As indicated in

the Methods section, it is important to mention that before removing the

halothane, we supplemented the anesthesia with ketamine.

We found evidence that isoflurane, an anesthetic agent with similar

pharmacological properties to halothane, is also capable of inducing strong

oscillations in the retinogeniculate responses (data not shown).

Taken together, these findings indicate that halogenated anesthetics have a

profound effect in the temporal patterning of single cell responses.

4.4 . NO-H A L O T H A N E C O N D I T I O N

In a series of experiments, data were obtained in the complete absence of

halothane (N= 2 cats). For this, recordings were made from the LGN under

ketamine, without immediate exposition to halothane (the ketamine-only

condition).

Tim

e (m

s)

0

80

-80

cgl03d11 4a-4b

cgl03d12 4a-4b

Coi

ncid

ence

s

0.2

0.0

Tim

e (m

s)

0

80

-80

Coi

ncid

ence

s

0.2

0.0

T ime

0.6%

KETAMINE

HALOTHANE

cgl03d04 4b

cgl03d04 4a

1aON-cell

LGN

ON-cell1b

LGN

2°

804010 120

804010 120

92 Hz

Frequency (Hz)

0.0

1.0

0.0

1.0

Figure 19. Oscillations are absent during ketamine anesthesia. Ketamine was given without previous exposition to halothane (ketamine-only condition). Notice that very strong synchronous oscillations appear after administration of halothane (0.6 %). Coherence is shown to the right. Oscillation frequency 92 Hz. Pair of ON-cells in the LGN. Overlapping RFs.


Figure 19 shows an example for responses in the LGN of two cells with

overlapping RFs. Synchronous oscillations that were completely absent from the

responses in the ketamine-only conditions, appeared strong as usual when

halothane is administered to the cat.

4.5 . RE C O R D I N G S I N T H E AWA K E C AT

Finally, in last series of experiments in one awake cat (shk) we could verify

whether fast retinal oscillations are present during a freely viewing condition.

As indicated in Figure 21, in absence of halothane the strong oscillations

in the LGN responses disappear.

Notably, in a series of sessions we were able to record LGN responses in

an awake cat, which was subsequently anesthetized with halothane, keeping the

same recording site (Figure 22). Oscillations were completely absent in the

awake condition and appeared strong as usual during the halothane anesthesia.

!

Time (ms) Frequency (Hz)Time804010 120

0

80

-80

shk005e01 4a, 4b, 4c, 4d, 4f

Coi

ncid

ence

s

0.2

0.0

86 Hz

0-800.0

0.1

0.0

11

80

0

80

-80

shk005b06 4a, 4b, 4c, 4d, 4f

Coi

ncid

ence

s

0.2

0.0

0-80 80 804010 1200.0

0.1

0.0

11

Tim

e (m

s)

0

80

-80

shk005b03 4a, 4b, 4c, 4d, 4f

Coi

ncid

ence

s

0.2

0.0

0-80 80 804010 1200.0

0.1

0.0

11

KETAMINEONLY

AWAKE

1.0%HALOTHANE

20°

(1a, 1b, 1c, 1d)

LGNON-ce l l s

Figure 20. In the awake cat, fast retinal oscillations are absent. Analysis was made for jointly responses of 4 ON-cells recorded simultaneously in the LGN, lamina A. Notice that the oscillations are completely absent from the responses for the awake cat and for the ketamine-only conditions. Following administration of halothane characteristically strong oscillations appear. Sliding window autocorrelation functions are shown to the left. Spectral analysis is shown to the right. Oscillation frequency 86 Hz. Overlapping RFs (plots are not shown).


!!

Time (ms) Frequency (Hz)Time804010 120

0

80

-80

shk005e01 4a, 4b, 4c, 4d, 4f

Coi

ncid

ence

s

0.1

0.0

72 Hz

0-500.0

0.3

0.0

11

50

Tim

e (m

s)

0

80

-80

shk005b03 4a, 4b, 4c, 4d, 4f

Coi

ncid

ence

s

0.1

0.0

0-50 50 804010 1200.0

0.3

0.0

11

(KETAMINE)

AWAKE

1.0%HALOTHANE

(1a, 1b) - (1c, 1d)

LGNON-ce l l s

(1a, 1b) - (1c, 1d)

LGNON-ce l l s

20°

Figure 21. Recordings in the LGN of an alert cat and during halothane anesthesia. Jointly responses from 2 pairs of ON-cells recorded simultaneously in lamina A1of the LGN. In this experiment, we were able to record LGN responses in an awake cat (shk), which was subsequently anesthetized with halothane (same recording site). Notice that exorbitantly strong oscillations appear following administration of halothane (1.0%). The central bin in the crosscorrelogram is equal zero because superimposed spikes were discarded in the spike sorting process. Coherence is shown to the right. Oscillation frequency 72 Hz. Overlapping RFs (plots are not shown). Correlograms are shown with a time lag ranging from -50 to 50 ms.

Time

0

80

-80

shk009i03 4a (3)

0.2

0.0

0

80

-80

shk009i01 4a (3)

0.2

0.0

Tim

e (m

s)

0

80

-80

shk009h04 4a (3)

Coi

ncid

ence

s

0.2

0.0

0.0

9.0

0.0

9.0

0.0

9.0

60 140Frequency (Hz)

100

120

75

AWAKESCREEN

Time

0

80

-80

shk011j01 (3a, 3b) (3)

0.2

0.0

0

80

-80

shk011j03 (3a, 3b) (3)

0.2

0.0

Tim

e (m

s)

0

80

-80

shk011j02 (3a, 3b) (3)

Coi

ncid

ence

s

0.2

0.0

0.0

9.0

0.0

9.0

0.0

9.0

60 140Frequency (Hz)

84

81

87

HALOTHANE1.0%

75 Hz

100 Hz

120 Hz

20° 20°

Figure 22. Entrainment of responses to the refresh of a CRT monitor display.


4.2 . ST I M U L U S E N T R A I N M E N T

In many recordings in the awake cat we have seen strong entrainment to

the periodic refresh of the monitor. Figure 22 compares the responses to

different CRT refresh frequencies (75 Hz, 100 Hz and 120 Hz) for the awake cat

and during halothane anesthesia. Observe that the recordings made in the awake

cat show strong entrainment with oscillation frequency coinciding precisely with

the refresh frequency, without any signs of decay. In the contrary, during the

anesthesia, strong oscillations are seen with about the same frequency (around

85 Hz). Surprisingly, these strong oscillatory patterns generated internally

overrides completely the oscillatory inputs (due to the refresh of the CRT

screen). Overall these findings show that responses in the retina of the cat during

natural conditions can be temporally precise, even without any oscillatory

patterning being generated internally.

!!!!!!!!!!!!!!!!


5. DI S C U S S I O N

The primary goal of this study was to test whether fast retinal oscillations

are present or not in conditions compatible with natural seeing. For this we

recorded directly from the retina of the anesthetized cat to study spiking

responses to dynamical stimuli, such as natural scenes movies and varying-size

stimuli (size- and luminance random walks). Finally, we obtained data from the

alert cat, removed from any influences of anesthetic agents.

It is well known that oscillations originated in the retina are transmitted

to the LGN and to the cortex (Doty and Kimura, 1963; Laufer and Verzeano,

1967; Arnett, 1975; Neuenschwander et al., 1996; Castelo-Branco et al., 1998). In

the study of Neuenschwander et al. (1996) synchronization of oscillatory

responses was observed for large distances in the retina (up to 20°), and could

be observed for LGN cells receiving inputs from the same eye, independent

whether the cells were located in the same LGN or in the LGNs of the two

hemispheres. Oscillatory patterns in the LGN can propagate to the cortex, as

demonstrated directly by simultaneous recordings from the retina, LGN and

areas A17 and A18 (Castelo-Branco et al., 1988).

An important aspect in the organization of the retinogeniculate system is

the reliability of information transfer. Retinal afferents to the LGN consist of

thick axons involving richly branched terminal arbors with boutons densely

distributed in terminal clusters (Sherman and Guillery, 2006). This organization

confers great robustness for the retinogeniculate transmission. Analysis of retinal

excitatory post-synaptic potentials (EPSPs) associated with LGN spike waveforms

(S-Potentials) shows that most LGN neurons have one retinal ganglion cell input

that accounts for nearly all LGN spikes sent to visual cortex (Sincich et al., 2007).

Thus, it is not surprising to see that the retinal fast oscillations are faithfully

transmitted through the retinogeniculate pathway up to the LGN (Figure 4).

It has been known that the size of simple stimuli modulates responses in

the early visual system (Barlow et al., 1954; Hubel and Wiesel, 1961). Small spots

of light flashed over the RF center evoke strong responses. However, when the

stimulus reaches the surround regions, the responses decreases. Therefore,


response levels are unlikely to be appropriate for encoding the size of the

stimulus.

An attractive alternative hypothesis could be that global characteristics

such as s ize and continuity are represented by a temporal code .

Neuenschwander and collaborators have shown in the LGN and the retina of the

cat that oscillation strength is correlated with stimulus size, using light rectangles

flashed over the RFs (Neuenschwander et al., 1999; Stephens et al., 2006). Our

study extended findings, showing that oscillations may encode the size of even

surfaces in natural images. As shown in Figure 18, oscillations in response to

natural scene stimuli appeared only for short epochs, when large connected

segments in the image (e.g., surfaces or the faces of a geometrical object)

covered the RFs. Qualitatively, we can see that the critical blob size for triggering

oscillations is very large. In this sense, encoding of size would be in place only

after a certain threshold is exceeded.

We found abundant evidence, with a variety of visual stimuli, that retinal

oscillations in the cat are dependent on halothane anesthesia, and are absent

during ketamina anesthesia or in natural conditions (Figures 18 to 21). Early

studies on temporal coding in the retinogeniculate system have used anesthetized

animals as experimental model, applying as anesthetic agent either sodium

thiopental (Doty and Kimura, 1963; Laufer and Verzeano, 1967; Reinagel and Reid,

2000; Butts et al., 2007; Desbordes et al., 2008), halothane (Neuenschwander et

al., 1996; Castelo-Branco et al., 1998) or isoflurane (Ito et al., 2010). Halogenated

anesthetics, such as halothane, are known to act directly on GABAA receptors

through an agonist effect by prolonging the decay of inhibitory postsynaptic

currents and increasing IPSP amplitudes (Li et al., 2000; Nishikawa and Maclver,

2000). Halothane is a general inhalation anesthetic used for induction and

maintenance of general anesthesia. It reduces the blood pressure and frequently

decreases the pulse rate and depresses respiration. The anesthetic also induces

muscle relaxation and reduces pains sensitivity by altering tissue excitability. It

does so by decreasing the extent of gap junction mediated cell-cell coupling and

altering the activity of the channels that underlie the action potential. Halothane

causes general anesthesia due to its actions on multiple ion channels, which

ultimately depresses nerve conduction, breathing, cardiac contractility and


reduces the consumption of oxygen up to 15%. Halothane was also shown to

facilitate evoked gamma in the rat visual cortex (Imas et al., 2004).

The gaseous anesthetic halothane have long been known to reduce gap

junction function (Rozental et al., 2000). The gap junctions or the electrical

synapses are known to mediate the retina’s ability to respond flexibly. Changes in

light intensity have been shown to regulate electrical synapses in at least three

places, between rods and cones, between horizontal cells, and between AII

amacrine cells (Kazumichi Shimizu and Mark Stopfer, 2013).

In this study, we show that isoflurane also evoked oscillatory responses

with the same temporal characteristics as halothane. Thus, our findings can be

generalized to halogenated anesthetics. This is important since these two agents

have been routinely used in many studies of temporal coding at various levels in

the sensory systems (e.g., Neuenschwander and Singer, 1996; Ito et al. 2010).

Finally, a few studies have reported that retinal oscillations occur also for

non-anesthetized and paralyzed cats and monkeys (after transpontine

transection) (Doty and Kimura, 1963), which is at odds with our results. In these

early experiments, however, the stimulus conditions were very different from

ours. Visual stimulation was generally made by illuminating the whole visual field

using light flashes (Doty and Kimura, 1963; Laufer and Verzeano, 1967). It remains

to be investigated whether whole-field stimulation (with a DC-light) are able to

entrain oscillations in the retina.


6. CO N C L U S I O N

In this study we show that fast retinal oscillations in the cat (recorded

both, directly from the retina and/ or the lateral geniculate nucleus) depend

strongly on halothane (or isoflurane) anesthesia, and are absent under natural

conditions, such as freely viewing natural scenes or artificial stimuli. These

findings raise serious doubts about the role of fast retinal oscillations on visual

processing. It is likely that fast rhythms in the retina arise from an imbalance

between excitation and inhibition produced by the anesthesia. Interestingly,

recordings from the awake cat demonstrate that retinal ganglion cell responses

may be entrained by fast periodic inputs as fast as 120 Hz (CRT monitor refresh

rate), indicating that the visual system of the cat is capable of representing very

fast events. Thus, although periodic rhythms are absent from retinal ganglion cell

responses under natural conditions, responses can be temporally precise.


7. RE F E R E N C E S

Ahissar, E., Sosnik, R., Haidarliu, S. (2000). Transformation from temporal to rate coding in a somatosensory thalamocortical pathway. Nature. 406, 302 – 306;

Alonso, J.M., Usrey, W.M., Reid, R.C. (1996). Precisely correlated firing in cells of the lateral geniculate. Nature. 383 (1996) 815–819;

Andolina, I.M., Jones, H.E., Wang, W., Sillito, A.M. (2007). Corticothalamic feedback enhances stimulus response precision in the visual system. PNAS. 104, 1685 – 1690;

Arai, I., Yamada, Y., Asaka, T., Tachibana, M. (2004). Light-Evoked Oscillatory Discharges in Retinal Ganglion Cells Are Generated by Rhythmic Synaptic Inputs. J. Neurophysiol. 92, 715 – 725;

Ariel, M., Daw, N.W., Rader, R.K. (1983). Rhythmicity in rabbit retinal ganglion cell responses. Vis. Re. 23, 1485 – 1493;

Arnett, D.W. (1975). Correlation analysis of units recorded in the cat dorsal lateral geniculate nucleus. Exp. Brain Res. 24 (1975) 111–130;

Barlow, H. B., Fitzhugh, R., Kuffler, S.W. (1954). Resting discharge and dark adaptation in the cat, J. Physiol. (Lond.), 25 28–9P;

Bishop, G.H. (1933). Fiber groups in the optic nerve. American Journal of Physiology. 106, 460 – 470;

Bishop, P.O., Levick, W.R., Williams, W.O. (1964). Statistical Analysis of the Dark Discharge of Lateral Geniculate Neurones. J. Physiol. 170: 598 - 612;

Butts, D.A., Weng, C., Jin, J., Yeh, C., Lesica, N.A., Alonso, J.M., Stanley, G.B. (2007). Temporal precision in the neural code and the timescales of natural vision. Nature. 449, 92 – 96;

Buzsáki, G. and Wang, X.J. (2012). Mechanisms of Gamma Oscillations. Annu. Rev. Neurosci. 35, 203 – 25;

Casagrande, V.A. and Kaas, J.H. (1994). The afferent, intrinsic, and efferent connections of primary visual cortex in primates. Cerebral Cortex. 10, 201 – 259;

Castelo-Branco, M., Goebel, R., Neuenschwander, S., Singer, W. (2000). Neural synchrony correlates with surface segregation rules. Nature. 405, 685 – 689;

Castelo-Branco, M., Neuenschwander, S., Singer, W. (1998). Synchronization of visual responses between the cortex, lateral geniculate nucleus, and retina in the anesthetized cat. The Journal of Neuroscience. 18, 6395 – 6410;

Dan, Y., Alonso, J.M., Usrey, W.M., Reid, R.C. (1998). Coding of visual information by precisely correlated spikes in the lateral geniculate nucleus, Nat Neurosci. 1, 501–507;

Desbordes, G., Jin, J., Weng, C., Lesica, N.A., Stanley, G.B., Alonso, J.M. (2008). Timing Precision in Population Coding of Natural Scenes in the Early Visual System. Plos Biology. 6, 2672 – 2682;


Doty, R.W., and Kimura, D.S. (1963). Oscillatory Potentials in the Visual System of Cats and Monkeys. J. Physiol. 168, 205 – 218;

Eckhorn, R., Bauer, R., Jordan, W., Brosch, M., Kruse, W., Munk, M., Reitboeck, H.J. (1988). Coherent Oscillations: A Mechanism of Feature Linking in the Visual Cortex? Biological Cybernetics. 60, 121 – 130;

Engel, A.K., Fries, P., Singer, W. (2001). Dynamic predictions: oscillations and synchronization in top-down processing. Nature. 2, 704 – 716;

Engel, A.K., König, P. and Singer, W. (1991). Direct physiological evidence for scene segmentation by temporal coding. Proc. Natl. Acad. Sci. USA 88, 9136 – 9140;

Enroth-Cugell, C., and Robson, J.G. (1966). The contrast sensitivity of retinal ganglion cells of the cat. J. Physiol. 187, 517 – 552;

Fries, P., Reynolds, J.H., Rorie, A.E., Desimone, R. (2001). Modulation of Oscillatory Neuronal Synchronization by Selective Visual Attention. Science. 291, 1560 – 1563;

Fries, P. (2009). Neuronal gamma-band synchronization as a fundamental process in cortical computation, Annu. Rev. Neurosci. 32, 209–224;

Fröhlich, F. W. (1914). Beiträge zur allgemeinen Physiologie der Sinnesorgane. Psychologische Physiologie Sinnesorganische II. Abteilung Sinnesphysiologische, 48, 28–164;

Gasser, H.S. and Erlanger, J. (1929). Role of fiber size in establishment of nerve block by pressure or cocaine. Amer. J. Physiol. 88, 581–591;

Gattass, R., Nascimento-Silva, S., Soares, J.G.M., Lima, B., Jansen, A.K., Diogo, A.C.M., Farias, M.F., Marcondes, M., Botelho, E.P., Mariani, O.S., Azzi, J., Fiorani, M. (2005). Cortical visual areas in monkeys: location, topography, connections, columns, plasticity and cortical dynamics. Phil. Trans. R. Soc. B. 360, 709 –731;

Gollisch, T. and Meister, M. (2008). Rapid Neural Coding in the Retina with Relative Spike Latencies. Science. 319: 1108 - 1111;

Gotch F. (1903). The time relations of the photo-electric changes in the eyeball of the frog. J Physiol. 29(4-5):388–410;

Gray, C.M. and Viana Di Prisco, G. (1997). Stimulus-dependent neuronal oscillations and local synchronization in striate cortex of the alert cat, J Neurosci. 17, 3239–3253;

Gray, C.M. and Singer, W. (1989). Stimulus-specific neuronal oscillations in orientation columns of cat visual cortex. Proc. Natl. Acad. Sci. USA. 86, 1698 – 1702;

Gray, C. M., König P., Engel A. K., Singer W. (1989). Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties, Nature. 338, 334–337;

Granit, R., Therman, P.O. (1935). Excitation and inhibition in the retina and in the optic nerve, J. Physiol. (Lond.). 83, 359–381;


Hasenstaub, A., Shu, Y., Haider, B., Kraushaar, U., Duque, A., McCormick. (2005). Inhibitory Postsynaptic Potentials Carry Synchronized Frequency Information in Active Cortical Networks. Neuron. 47, 423 – 435;

Haslinger, R., Pipa, G., Lima, B., Singer, W., Brown, E.N., Neuenschwander, S. (2012). Context Matters: The Illusive Simplicity of Macaque V1 PLoS ONE. 7 (2012) e39699;

He, D.S., Burt, J.M. (2000). Mechanism and Selectivity of the Effects of Halothane on Gap Junction Channel Function. Circulation Research. 86, 104 – 109;

Herculano-Houzel, S., Munk, M. H., Neuenschwander, S., Singer, W. (1999) Precisely synchronized oscillatory firing patterns require electroencephalographic activation, Journal of Neuroscience. 19, 3992–4010;

Hoffmann, K.P., Stone, J., Sherman, M.S. (1972). Relay of Receptive-Field Properties in Dorsal Lateral Geniculate Nucleus of the Cat. J. Neurophysiol. 35, 518 – 521;

Hormuzdi, S.G., Filippov, M.A., Mitropoulou, G., Monyer, H., Bruzzone, R. (2004). Electrical synapses: a dynamic signaling system that shapes the activity of neuronal networks. Biochimica et Biophysica Acta. 1662, 113 – 137;

Hubel, D.H., Wiesel, T.N. (1961). Integrative action in the cat’s lateral geniculate body. J. Physiol. 155, 385 – 398;

Hubel, D.H., Wiesel, T.N. (1962). Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. J. Physiol. 160, 106 – 154;

Hughes, T.E., Grünert, U., Karten, H.J. (1991). Receptors in the retina of the cat: an immunohistochemical study of wholemounts, sections, and dissociated cells. Vis. Neurosci. 6, 229 – 238;

Imas, O.A., Ropella, K.M., Wood, J.D., Hudetz, A.G. (2004). Halothane augments event-related γ oscillations in rat visual cortex. Neuroscience. 123, 269 – 278;

Ishikane, H., Gangi, M., Honda, S., Tachibana, M. (2005). Synchronized retinal oscillations encode essential information for escape behavior in frogs. Nature Neuroscience. 8, 1087 – 1095;

Ishikane, H., Kawana, A., Tachibana, M. (1999). Short- and long-range synchronous activities in dimming detectors of the frog retina. Visual Neuroscience. 16, 1001 – 1014;

Ito, H., Maldonado, P.E., Gray, C.M. (2010). Dynamics of Stimulus-Evoked Timing Correlations in the Cat Lateral Geniculate Nucleus. J. Neurophysiol. 104, 3276 – 3292;

Kirschfeld, K. (1992). Oscillations in the insect brain: do they correspond to the cortical gamma-waves of vertebrates? Proc. Natl. Acad. Sci. USA. 15, 4764 – 4768;

Koepsell, K., Wang, X., Vaingankar, V., Wei, Y., Wang, Q., Rathbun, D.L., Usrey, W.M., Hirsch, J.A., Sommer, F. (2009). Retinal oscillations carry visual information to cortex. Frontiers in systems neuroscience. 3, 1 – 18;

Koepsell, K., Wang, X., Hirsch, J.A., Sommer, F.T. (2010). Exploring the function of neural oscillations in early sensory systems. Frontiers in Neuroscience. 4, 53 – 61;


König, P. (1994) A method for the quantification of synchrony and oscillatory properties of neuronal activity. J Neurosci Methods 54:31–37;

Kotani, N. and Akaike, N. (2013). The effects of volatile anesthetics on synaptic and extrasynaptic GABA-induced neurotransmission. Brain Research Bulletin. 93, 69 – 79;

Kreiter, A.K. and Singer W. (1996). Stimulus-dependent synchronization of neuronal responses in the visual cortex of the awake macaque monkey, J Neurosci. 16 2381–2396;

Krasowski, M.D. and Harrison, N.L. (1999). General anaesthetic actions on ligand-gated ion channels. Cell Mol Life Sci. 55, 1278 – 1303;

Kratz, K.E., Webb, S.V., Sherman, S.M. (1978). Electrophysiological classification of X- and Y- cells in the cat’s lateral geniculate nucleus. Vision Res. 18, 489 – 492;

Kuffler, S. (1953). Discharge patterns and functional organization of mammalian retina, J Neurophysiol. 16 37–68;

Laufer, M. and Verzeano, M. (1967). Periodic activity in the visual system of the cat. Vision Res. 7, 215 – 229;

Laurent, G. (2002). Olfactory networks dynamics and the coding of multidimensional signals. Nature. 3, 884 – 895;

Li, X., Czajkowski, C., Pearce, R.A. (2000). Rapid and Direct Modulation of Receptors by Halothane. Anestheology. 92, 1366 – 1375;

Lima, B., Singer, W., Chen, N., Neuenschwander, S. (2009). Synchronization Dynamics in Response to Plaid Stimuli in Monkey V1. Cerebral Cortex. 20, 1556 – 1573;

Lima, B., Singer, W., Neuenschwander, S. (2011). Gamma Responses Correlate with Temporal Expectation in Monkey Primary Visual Cortex. The Journal of Neuroscience. 31, 15919 – 15931;

Livingstone, M. and Hubel, D. (1988). Segregation of Form, Color, Movement, and Depth: Anatomy, Physiology, and Perception. Science. 240, 740 – 749;

Merker, B. (2013). Activation, not cognition, is the functional key to cortical gamma oscillations. Neuroscience and Biobehavioral Reviews. 37, 401 – 417;

Munk, M. H., Roelfsema, P. R., König, P., Engel, A. K., Singer, W. (1996). Role of reticular activation in the modulation of intracortical synchronization, Science. 272, 271–274;

Nassi, J.J., Callaway, E.M. (2009). Parallel Processing Strategies of the Primate Visual System. Nature Rev. Neurosci. 10, 360 – 372;

Neuenschwander, S., Castelo-Branco, M., Baron, J., Singer, W. (2002). Feed-forward synchronization: propagation of temporal patterns along the retinothalamocortical pathway. Phil. Trans. R. Soc. Lond. B. 357, 1869 – 1876;

Neuenschwander, S., Castelo-Branco, M., Singer, W. (1999). Synchronous oscillations in the cat retina. Vision Research. 39, 2485 – 2497;

Neuenschwander, S., Singer, W. (1996). Long-range synchronization of oscillatory light responses in the cat retina and lateral geniculate nucleus. Nature. 379, 728 – 733;


Nishikawa, K. and Maclver, M.B. (2000). Membrane and Synaptic Actions of Halothane on Rat Hippocampal Pyramidal Neurons and Inhibitory Interneurons. The Journal of Neuroscience. 20, 5915 – 5923;

Palanca, B.J.A. and DeAngelis, B.J.A. (2005). Does neuronal synchrony underlie visual feature grouping? Neuron. 46, 333–346;

Pipa, G., Chen, Z., Neuenschwander, S., Lima, B., Brown, E.N. (2012). Mapping of Visual Receptive Fields by Tomographic Reconstruction, Neural Computation. 24, 2543–2578;

Press. W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P. (1986). Numerical Recipes in C – The Art of Scientific Computing. Cambridge University Press;

Quiroga, R.Q., Nadasdy, Z., Ben-Shaul, Y. (2004). Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering. Neural Computation. 16, 1661 – 1687;

Reich, D.S., Victor, J.D., Knight, B.W., Ozaki, T., Kaplan, W. (1997). Response Variability and Timing Precision of Neuronal Spike Trains in Vivo. J. Neurophysiol. 77, 2836 – 2841;

Reinagel, P. and Reid, C. (2000). Temporal Coding of Visual Information in the Thalamus. The Journal of Neuroscience. 20, 5392 – 5400;

Roberts, W.M. and Rutherford, M.A. (2008). Linear and nonlinear processing in hair cells. The Journal of Experimental Biology. 211, 1775 – 1780;

Roelfsema, P., Lamme, V.A.F., Spekreijse, H. (2004). Synchrony and covariation of firing rates in the primary visual cortex during contour grouping. Nature Neuroscience. 7, 982 – 991;

Rosa, M.G.P. and Tweedale, R. (2005). Brain maps, great and small: lessons from comparative studies of primate visual cortical organization. Phil.Trans.R.Soc.B. 360, 665 – 691;

Saito, H.A. (2004). Morphology of physiologically identified X-, Y-, and W-type retinal ganglion cells of the cat. Journal of Comparative Neurology. 221, 279 – 288;

Sherman, S.M. and Guillery, R.W. (2006). Exploring the thalamus and its role in cortical function. Sec. Edition. The MIT Press. Cambridge, Massachusetts. London, England;

Sherman, S.M., Guillery, R.W. (2002). The role of the thalamus in the flow of information to the cortex. Phil. Trans. R. Soc. Lond. B. 357, 1695 – 1708;

Sillito, A.M., Cudeiro, J., Jones, H.E. (2006). Always returning: feedback and sensory processing in visual córtex and thalamus. Trends in Neurosciences. 29, 307 – 316;

Sincich, L.C., Adams, D.L., Economides, J.R., Horton J.C. (2007). Transmission of spike trains at the retinogeniculate synapse, J Neurosci. 27, 2683–2692;

Singer, W. (1999). Neuronal Synchrony: A versatile Code for the Definition of Relations? Neuron. 24, 49-65;


Söhl, G., Maxeiner, S., Willecke, K. (2005). Expression and functions of neuronal gap junctions. Nature. 6, 191 – 200;

Stephens, G.J., Neuenschwander, S., George, J.S., Singer, W., Kenyon, G.T. (2006). See globally, spike locally: oscillations in a retinal model encode large visual features. Biol. Cybern. 95, 327 – 348;

Steriade, M., Contreras, D., Amzica, F., Timofeev, I. (1996). Synchronization of Fast (30 – 40 Hz) Spontaneous Oscillations in Intrathalamic and Thalamocortical Networks. The Journal of Neuroscience. 16, 2788 – 2808;

Stanley, G.B., Jin, J., Wang, Y., Desbordes, G., Wang, Q., Black, M.J. , Alonso, J-M (2012). Visual orientation and directional selectivity through thalamic synchrony. Journal of Neuroscience, 32 (2012) 9073–9088.

Stone J. (1983). Parallel processing in the Visual System. New York: Plenum Press; Thiele, A. and Stoner, G. (2003). Neuronal synchrony does not correlate with motion

coherence in cortical area MT. Nature. 421, 366 – 370; Tolhurst, D.J. (1973). Separate channels for the analysis of the shape and the

movement of a moving visual stimulus. J. Physiol. 231, 385 – 402; Uhlhaas, P.J., Pipa, G., Neuenschwander, S., Wibral, M., Singer, W. (2011). A new

look at gamma? High- (>60 Hz) γ-band activity in cortical networks: Function, mechanisms and impairment. Progress in Biophysics and Molecular Biology. 105, 14-28;

Usrey, M.W. (2002). Spike timing and visual processing in the retinogeniculocortical pathway. Phil. Trans. R. Soc. Lond. B. 357, 1729 – 1737;

Usrey, W.M., Reppas, J.B., Reid, R.C. (1998). Paired-spike interactions and synaptic efficacy of retinal inputs to the thalamus. Nature. 395, 384 – 387;

Vardi, N., Masarachia, P., Sterling, P. (1992). Immunoreactivity to Receptor in the Outer Plexiform Layer of the Cat Retina. The Journal of Comparative Neurology. 320, 394 – 397;

Von der Malsburg, C. (1981). The Correlation Theory of Brain Function. Internal Report 81-2, Dept. of Neurobiology, Max-Planck Institute for Biophysical Chemistry, Göttingern, Germany;

Wang, X. and Buzsáki, G. (1996). Gamma Oscillation by Synaptic Inhibition in a Hippocampal Interneuronal Network Model. The Journal of Neuroscience. 16, 6402 – 6413;

Wentlandt, K., Samoilova, M., Carlen, P.L., El Beheiry, H. (2006). General Anesthetics Inhibit Gap Junction Communication in Cultured Organotypic Hippocampal Slices. Anesth. Analg. 102, 1692 – 1698;

Womelsdorf, T., Schoffelen, J.M., Oostenveld, R., Singer, W., Desimone, R., Engel, A., Fries, P. (2007). Modulation of Neuronal Interactions Through Neuronal Synchronization. Science. 316, 1609 – 1612;

Wunderle, T., Eriksson, D., Schmidt, K.E. (2013). Multiplicative Mechanism of Lateral Interactions Revealed by Controlling Interhemispheric Input. Cerebral Cortex. 23, 900 – 912;


Yeh, C., Stoelzel, C.R., Alonso, J.M. (2003). Two different types of Y cells in the cat Lateral Geniculate Nucleus. J. Neurophysiol. 90, 1852 – 1864.


8. TA B L E S

Table 1 . Summary of number of recording sites and all stimulus protocols used in this study.C

atR

eco

rdin

g A

rea

Tecn

iqu

eA

nes

thes

iaM

apR

FC

ircl

eSi

zeA

nn

ula

rMas

kG

rati

ngs

Size

Wal

kLu

min

ance

Wal

kLa

bP

anM

edik

amen

tTo

tal

Nr.

of

MU

As

Rec

ord

ing

Site

s

Shak

iLG

NSe

mi-C

hron

icAw

ake

71

641

73

20

81

11

Shak

iLG

NSe

mi-C

hron

icKe

tam

ine

only

12

14

12

3

Shak

iLG

NSe

mi-C

hron

icHa

loth

ane

+ Ke

tam

ine

41

101

52

61

8

Nala

LGN

Sem

i-Chr

onic

Keta

min

e on

ly5

15

11

23

7

Nala

LGN

Acut

eKe

tam

ine

24

56

17

51

21

Nala

LGN

Acut

eHa

loth

ane+

Ke

tam

ine

110

14

32

52

67

81

2

Nala

LGN

Acut

eHa

loth

ane

42

34

13

39

12

Nala

Retin

aAc

ute

Keta

min

e4

52

11

24

13

Nala

Retin

aAc

ute

Halo

than

e+

Keta

min

e1

93

22

32

04

61

2

Nala

Retin

aAc

ute

Halo

than

e4

23

41

32

91

1

cgl0

7LG

NAc

ute

Halo

than

e1

22

11

77

3

cgl0

7LG

NAc

ute

Halo

than

e +

Keta

min

e1

23

31

cgl0

7LG

NAc

ute

Keta

min

e1

22

55

1

cgl0

7Re

tina

Acut

eHa

loth

ane

12

21

17

71

cgl0

7Re

tina

Acut

eHa

loth

ane

+ Ke

tam

ine

12

33

1

cgl0

7Re

tina

Acut

eKe

tam

ine

12

25

52

cgl0

4LG

NAc

ute

Halo

than

e18

410

46

42

12

92

4

cgl0

4LG

NAc

ute

Halo

than

e +

Keta

min

e*1

11

13

33

cgl0

4LG

NAc

ute

Keta

min

e*2

26

3

cgl0

3LG

NAc

ute

Halo

than

e9

76

72

98

71

5

cgl0

3LG

NAc

ute

Keta

min

e3

25

15

6

cgl0

3LG

NAc

ute

Isoflu

rane

11

26

3

*was

hout

pro

coto

ls


9. F I N A N C I A L S U P P O RT

During this work Giovanne Rosso was supported by CAPES (Master‘s

Degree Scholarship). Our laboratory (Vislab, ICe-UFRN) was supported by the

Federal University of Rio Grande do Norte-UFRN, CNPq (BMBF-CNPq 490127/

2011-8, CNPq Universal 478060/2012-2, CNPq PQ 308190/2012-2) and by a

collaboration between the Brain Institute-UFRN and the Max-Planck Institute for

Brain Research, Frankfurt.

This thesis was produced as part of the activities of FAPESP Center for

Neuromathematics (FAPESP 2013/ 07699-0, São Paulo Research Foundation).

!


!!

10. AC K N O W L E D G M E N T S

Many thanks to my advisor Sergio Neuenschwander. Things may go wrong,

but as he always says „we need to fight“. After all, there is no free lunch. Sergio is

a great experimentalist, not only because of his knowledge and skills, but mostly

because of the way he lives and loves science. Thanks for the patience and for

believing on me, whenever I hesitated.

I also wish to thank Kerstin Schmidt, Jerome Baron, Bruss Lima, Ed

Tehovinik, Luiz Lana and Johanna Klon-Lipok for the suggestions, discussions,

ideas and support given during this scientific journey.

Thanks to Dr. med. vet. Josy Pontes for help and care with the cats. Many

thanks to Heitor de Oliveira for his wonderful 3D renderings and design of

mechanical stuff essential to this work. Special thanks to Witilla for her care and

commitment in the animal house. Thanks also to Marilene, Monik, Roseleide and

Eronildo for help in many occasions.

I also wish to thank Katia-Simone Rocha, Stephany Campanelli, Dardo

Ferreiro for the discussions, criticisms and friendship.

Many thanks to Prof. Jerome Baron and Prof. Claudio Queiroz for kindly

accepting to evaluate my thesis work, and Akaline Araújo for her secretary work

at the PG-Neuro, UFRN.

Finally I wish to thanks my family for continuous support. Thank you Dad

and Mom for the attention and love. Guilherme, my brother, thank you for the

dedication and commitment. Vô Clarindo, thanks for being always a great father,

grandfather and a friend. I wish you could be here.

!!!!!


!!!!!!!!!!!!!!!!!

INSTITUTO DO CÉREBRO - UFRN

AV. NASCIMENTO DE CASTRO 2155

59056-450 NATAL

HTTP://WWW.NEURO.UFRN.BR/RESEARCH/GROUPS/9

VISLAB

http://www.neuro.ufrn.br/research/groups/9

Documents

GIOVANNE ROSSO - Federal University of Rio Grande do … · universidade federal do rio grande do norte! programa de pÓs-graduaÇÃo em neurociÊncias! "! !!! do fast retinal oscillations