Just as I thought yesterday that it is possible to use pRF for long-range interaction, Doumolin (531.9) gave a talk this morning doing exactly that. They did put much promoting this technique.
However, I think Brewer did a better job (580.7-580.8). Her group associate pRF and retinotopic mapping and showed the pinwheel structure all the way down to TO.
Zotto et al. (581.22) showed that N190 is sensitive to the attractiveness of female body. I would say the difference is between naked and dressed body, not attractiveness.
Oh! And monkey has inversion effect for faces as well (581.23). And there is a study on the other-species, not other-race effect in face recognition. Fun!
Color vision symposium
620.4 Single opponent cells in V1 are not orientation tuned, but double opponent cells are. So, the excitatory and inhibitory regions of double opponent cells are not concentric but elongated and situated side by side.
620.5 Horwitz use staircase procedure to find the isoresponse contour of a neuron in a color space. Start with a low contrast, increase it until the response of the neuron hits a criteria, then, decrease the contrast. This procedure go back and forth for several reversals. Then, it is the isoresponse for that color direction. The data is then fit with qudratic planes.
The shape of the quadratic plane can be explained by a hierarchical model of neurons. The second order neuron responds only if two orthogonal (in terms of sensitivity plane) first order neurons fires.
2010年11月16日 星期二
2010年11月15日 星期一
SFN2010 Note #2
326.6 Golomb & Kanwisher measured BOLD activation to images varied either in category or in position. They then used multivoxel classification technique to analyze classification performance in a bunch of ROIs. They showed that not only the early and dorsal visual areas coded position more than category and the ventral area coded categories, but also all the position coding are retinotopic, not spatiotopic.
325.9-11 Doumolin group is really working hard to sell their pRF (population receptive field) story. I can image how uscan eful it is if one map the RF of each individual voxel. With that, we can really talk about lateral interaction in many brain areas and to investigate the different response properties of these areas with one experiment.
However, in a couple hours later, LI & Freeman (483.8) showed the metabolic coupling of center and surround. That is, pRF may be influenced by surrounds by metabolic factors. Hence, it's accuracy is remain questionable.
Petro et al. (393.15) tried to measure the diagnostic features for expression discrimination with fMRI. They showed a face for expression discrimination task, then, use a checkerboard to identify the retinotopic areas corresponding to eyes and mouths. They showed that the mouth and the eye areas in V1 does not tell expression well, but the rest of V1 does.
This study itself is not well executed but the idea is nice. We may borrow this idea to do some good things.
325.9-11 Doumolin group is really working hard to sell their pRF (population receptive field) story. I can image how uscan eful it is if one map the RF of each individual voxel. With that, we can really talk about lateral interaction in many brain areas and to investigate the different response properties of these areas with one experiment.
However, in a couple hours later, LI & Freeman (483.8) showed the metabolic coupling of center and surround. That is, pRF may be influenced by surrounds by metabolic factors. Hence, it's accuracy is remain questionable.
Petro et al. (393.15) tried to measure the diagnostic features for expression discrimination with fMRI. They showed a face for expression discrimination task, then, use a checkerboard to identify the retinotopic areas corresponding to eyes and mouths. They showed that the mouth and the eye areas in V1 does not tell expression well, but the rest of V1 does.
This study itself is not well executed but the idea is nice. We may borrow this idea to do some good things.
2010年11月14日 星期日
SFN2010 Note #1
126.1 Rubin & Miller did a good job on surround integration. The contrast-size interaction itself is not that new. However, with other data, the authors do provide a good understanding of the surround effect.
126.3 So, the orientation tuning of visual cortical neurons can be derived from Moire interference on retinal input. This is too good to be true.
128.10 Fedorenko & Kanwisher showed that there is no language specific areas. Any know language areas also showed activation to a wide array of different cognitive tasks.
I am convinced that there is no specific language areas, at least no reading areas. However, reviewers complained every single time when I made a statement like that. This study is more comprehensive than what I did so far. I expect this study, once published, will have much impact and arouse a heated debate in psycho- and neuro-linguistic research.
226.1 Lewis et al. used resting state function connectivity to show that the normal control and the amblyopic connectivity is similar between different visual areas but different in self connectivity with areas in V1 to V3.
226.5 Tanifuji et al. use SVM on single cell recording data from almost 40 sits. They showed that the structure cluster is similar to functional cluster when compared faces with non-faces. However, to tell human faces from monkey faces, features from both face and non-face regions are required.
226.7 Saygin et al. suggested that the diffusion connectivity to the FFA would be different from the rest of the fusiform. Therefore, it is possible to identify FFA from the rest of the FG by classifying connection from other brain areas to FG against other FG areas.
I wonder whether this paradigm would work for VWFA as well.
226.9 VWFA respond to lumiannce-defined, contour defined and motion defined words. HMT responds only to motion defined words. However hMT is necessary for motion defined words
126.3 So, the orientation tuning of visual cortical neurons can be derived from Moire interference on retinal input. This is too good to be true.
128.10 Fedorenko & Kanwisher showed that there is no language specific areas. Any know language areas also showed activation to a wide array of different cognitive tasks.
I am convinced that there is no specific language areas, at least no reading areas. However, reviewers complained every single time when I made a statement like that. This study is more comprehensive than what I did so far. I expect this study, once published, will have much impact and arouse a heated debate in psycho- and neuro-linguistic research.
226.1 Lewis et al. used resting state function connectivity to show that the normal control and the amblyopic connectivity is similar between different visual areas but different in self connectivity with areas in V1 to V3.
226.5 Tanifuji et al. use SVM on single cell recording data from almost 40 sits. They showed that the structure cluster is similar to functional cluster when compared faces with non-faces. However, to tell human faces from monkey faces, features from both face and non-face regions are required.
226.7 Saygin et al. suggested that the diffusion connectivity to the FFA would be different from the rest of the fusiform. Therefore, it is possible to identify FFA from the rest of the FG by classifying connection from other brain areas to FG against other FG areas.
I wonder whether this paradigm would work for VWFA as well.
226.9 VWFA respond to lumiannce-defined, contour defined and motion defined words. HMT responds only to motion defined words. However hMT is necessary for motion defined words
訂閱：
文章 (Atom)