WEKO3
アイテム
Contribution of neurons in monkey parietal cortex to a visual grouping
https://ir.soken.ac.jp/records/2180
https://ir.soken.ac.jp/records/218062dc9e1f-0363-4379-a3aa-bc288223ec4a
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
要旨・審査要旨 (344.3 kB)
|
||
|
本文 (3.4 MB)
|
| Item type | 学位論文 / Thesis or Dissertation(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2011-06-03 | |||||
| タイトル | ||||||
| タイトル | Contribution of neurons in monkey parietal cortex to a visual grouping | |||||
| タイトル | ||||||
| タイトル | Contribution of neurons in monkey parietal cortex to a visual grouping | |||||
| 言語 | en | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_46ec | |||||
| 資源タイプ | thesis | |||||
| 著者名 |
横井, 功
× 横井, 功 |
|||||
| フリガナ |
ヨコイ, イサオ
× ヨコイ, イサオ |
|||||
| 著者 |
YOKOI, Isao
× YOKOI, Isao |
|||||
| 学位授与機関 | ||||||
| 学位授与機関名 | 総合研究大学院大学 | |||||
| 学位名 | ||||||
| 学位名 | 博士（理学） | |||||
| 学位記番号 | ||||||
| 内容記述タイプ | Other | |||||
| 内容記述 | 総研大乙第207号 | |||||
| 研究科 | ||||||
| 値 | 生命科学研究科 | |||||
| 専攻 | ||||||
| 値 | 20 生理科学専攻 | |||||
| 学位授与年月日 | ||||||
| 学位授与年月日 | 2010-09-30 | |||||
| 学位授与年度 | ||||||
| 値 | 2010 | |||||
| 要旨 | ||||||
| 内容記述タイプ | Other | |||||
| 内容記述 | <B>Introduction</B> Visual grouping is an essential component of the visual perception of objects. It is the process by which multiple discrete elements are bound into a single object. For example, if multiple dots with the same color are arranged along a straight line, these dots are grouped together and recognized as a single linear object. This grouping is caused by bottom-up factors such as similarity and continuity of the dots. It is known that visual grouping is also affected by top-down factors such as prior knowledge and past experience with the objects. Neurological observations made in human patients and in fMRI studies of healthy human subjects suggest that the posterior parietal cortex plays a key role in visual grouping. It remains unknown, however, how parietal cortex are involved in visual grouping. <b>Relationship between neural responses and visual grouping in the monkey parietal cortex</b> To investigate the neuronal mechanisms underlying visual grouping, we designed a grouping detection task controlled by top-down attention, and performed extra-cellular single unit recording from lateral bank of intra-parietal sulcus (L-IPS) while the task was being performed by monkeys. The visual stimuli consisted of multiple discrete dots, and the monkeys were required to detect the target defined by specific arrangements of the dots. In addition, we manipulated the monkeys’ attention to the grouping of the elements, and examined the effect of attention on the neuronal responses. The visual stimuli were composed of 5 square black or white dots (1.2 deg at the edge) arranged in a cross. A total of 20 types of visual stimuli composed of different arrangements of dots were prepared. In four of the 20 stimuli, three dots with the same contrast (either black or white) were aligned either horizontally or vertically and served as the target. The remaining 16 stimuli were non-targets. The target stimuli were characterized by two visual features: the orientation of the three dots with the same contrast that was either horizontal or vertical (target orientation) and the contrast of three-aligned dots, which was either white or black (target contrast). Visual stimuli were presented multiple times in one trial. The monkeys made behavioral response via a lever, and had to release the lever within 600 ms after the onset of the target to obtain a liquid reward. The monkeys performed the detection task while their attention was directed towards a particular orientation. The attended orientation was controlled either by a visual cue or by a biased block design. We recorded the activities of 107 single neurons in the L-IPS while two monkeys performed a grouping detection task. We found that L-IPS neurons selectively responded to the visual stimulus, and a majority of neurons exhibited stronger selectivity for the target orientation than the target contrast. This orientation selectivity was enhanced when the target orientation matched the attended orientation. Moreover, the orientation-selective responses correlated with the monkeys' behavior. These results suggest that L-IPS neurons play important roles in the visual grouping and detection of objects comprised of discrete elements. <b>Activities of different cell classes in visual grouping</b> Although it is known that there are two functional classes of cortical neurons, excitatory pyramidal neurons and inhibitory interneurons, it remains largely unknown how these two classes contribute to visual perception and cognition. Recently, several attempts have been made to classify extracellularly recorded neurons according to known differences in the waveforms of their action potentials (e.g., Mitchell et al., 2007, Neuron). These studies suggest that classification of neuron type will provide valuable new information that could be crucial to understanding neural processing within local circuits in the cerebral cortex. In order to examine how different classes of neurons are involved in visual grouping, we classified recorded neurons according to the waveforms of their action potentials, and compared the response properties of classified neurons. We found that putative pyramidal neurons, which had broader action potentials, exhibited selectivity for the target orientation, and the selectivity was enhanced by attention. By contrast, putative inhibitory neurons, which had narrower action potentials, did not exhibit such selectivity or enhancement. Instead interneurons responded more strongly to the target stimuli than to the non-targets, regardless of the orientation of the target. These results suggest that different classes of parietal neurons contribute differently to the visual grouping of discrete elements. <b>Feedback projection linking the visual fields surrounding the blind spot</b> Classification of L-IPS neurons showed that pyramidal neurons exhibited selectivity for the target stimulus, and clearly indicates that L-IPS neurons signal information about the grouped stimulus to other cortical areas. Neurons in L-IPS may provide feedback signals and affect the activity related to visual grouping in the early visual area. However, no study has explored in detail the feedback projection related to the visual grouping. In an attempt to study the contribution of feedback projection on visual grouping, we examined whether there is an anatomical basis for integration of visual signals from both sides of blind spot (BS) by cortico-geniculate feedback neurons in V1. The blind spot is the region in the visual field that corresponds to the optic disk in the retina. No visual information exists in the blind spot because there are no photoreceptors within the optic disk. Nonetheless, we perceive color and/or patterns there that are the same as in the surrounding visual field. This phenomenon is known as perceptual filling-in, and closely related to the visual grouping. Neural mechanisms under perceptual filling-in at the blind spot has been examined in detail, and this provides a good physiological model to investigate the anatomical basis for integration of visual signals related to visual grouping. We recorded neuronal activity from V1 of a cat and mapped the receptive fields of V1 neurons. After identifying the blind spot region in V1, we inserted a glass micropipette (tip diameter, 25 μm) filled with biotinylated dextran amine (BDA) into a location adjacent to the blind spot region in V1, and injected BDA by iontophoresis. BDA labeled axons were traced around the neuron-free gap in layer A of LGN. When this has been finished in sections without counter-staining, we removed the cover glass of the sections and stained sections with cresyl violet for the reconstruction of the entire LGN. We observed that numerous axons traverse the neuron-free gap that retinotopically corresponds to BS within LGN. This indicates that visual signals from one side of BS are conveyed to the opposite side via a feedback connection. Cortico-geniculate feedback projection may integrate visual signals from around BS and contribute to perceptual filling-in at BS. <B>Conclusion</B> We recorded neuronal activities in L-IPS while monkeys performed a grouping detection task. We found that pyramidal neurons in L-IPS exhibited selectivity for the orientation of the target, and this selectivity was enhanced by attention to a particular target orientation. This result indicates that L-IPS neurons signal information about the grouped stimulus to other cortical areas. In the anatomical experiment, we found the feedback connection linking the visual fields surrounding the blind spot, which may be involved in the integration or interaction of visual information present from separate locations within the visual field. Neuronal activity in V1 is modulated by the presentation of visual stimuli in the receptive field surround, and it has been suggested that this contextual modulation is related to the visual grouping (Gilbert et al., 2000, Vision Res). Presumably, L-IPS neurons provide feedback signals to the early visual areas and facilitate visual grouping by way of the contextual modulation there. These results provide the first physiological evidence that L-IPS neurons make an important contribution to visual grouping by combining visual and attentional signals to bind discrete visual elements. A recurrent circuit between the L-IPS and early visual areas may be critical for visual grouping through the interchange of feedforward and feedback signals. |
|||||
| 所蔵 | ||||||
| 値 | 有 | |||||
