Results…………………………………………………………………………….68
Conclusion………………………………………………………………………..88
Figures……………………………………………………………………………90

Figure1. Stimulus set …………………………………….……………..….90
Figure 2. Different noise levels of an exemplar stimulus ……………….…91
Figure 3. Passive task……………………………………………………….92
Figure 4. Active task (two-alternative forced-choice body/object categorization…………………………………………………………….…93
Figure 5. Monkeys’ performance in body/object categorization task. …….95
Figure 6. The pattern of performance decline as a function of noise level was reverse for bodies and objects …………………………………………….. 96
Figure 7. Monkeys’ performance in body/object categorization task for subcategories …………………………………………………………..…..97
Figure 8 .Performance decline between adjacent signal levels in subcategories of bodies and objects ……………………………….………98
Figure 9. Behavioral d́ (d́ = Z “hit rate” – Z “false alarm”) in different visual signals ……………………………………………………………….…….. 99
Figure 10. Cumulative d́ in signal level of 90 ……………………….……100
Figure 11. Reaction time in different signal conditions in correct and wrong trials……………………………………………………………………….101
Figure 12. Reaction time in subcategory level……………………………102
Figure 13. Relation between reaction time and performance in different signal levels…………………………………………………………….…103
Figure 14. Mean number of microsaccades in different noise levels……..104
Figure 15. Mean number of microsaccades in correct and wrong trials of different signal levels……………………………………………………..105
Figure 16. Reaction time in trials with and without microsaccades in different signal levels…………………………………………………..…106
Figure 17. Normalized mean firing rate of body cells across different visual signals and behavioral conditions…………………………………….…..107
Figure 18. Normalized mean firing rate of non-body cells across different visual signals and behavioral conditions…………………………………..109
Figure 19.Response modulation index (RMI) as a function of task difficulty……………………………………………………………….….111
Figure 20. Mean response modulation onset across body image signal levels in body cells’ correct trials…………………………………………….….112
Figure 21. Attentional enhancement of IT cells’ body-object discriminability (d’) was observed only in correct trials and degree of enhancement depended on task difficulty………………………………………………………..…114
Figure 22. Mean d’ modulation in correct (blue) and wrong (black) compared to passive condition in body cells…………………………………………116

Figure 23. Mean d’ modulation in correct (blue) and wrong (black) compared to passive condition in non-body cells……………………………………117
Figure 24. Temporal pattern of baseline firing rate modulation in active compared to passive condition…………………………………………….118
Figure 25. Temporal pattern of p-values of t-tests measuring significant increase of baseline rate in active compared to passive condition……..…119
Figure 26. Frequency distribution of proportion of HBTs in body (top) and non-body (bottom) cells during active task…………………………….…120
Figure 27. Baseline dependent enhancement of body and suppression of non-body cells’ responses to presentation of body images in correct condition……………………………………………………………….….121
Figure 28. Baseline dependent enhancement of body and suppression of non-body cells’ responses to presentation of body images in wrong condition………………………………………………………………..…122
Figure 29. Temporal dynamic of body and non-body cells’ RMI to presentation of body images in correct and wrong conditions for HBTs………………………………………………………………………123
Figure 30. Temporal dynamic of body and non-body cells’ RMI to presentation of body images in correct and wrong conditions for LBTs…124
Figure 31. P-values of t-tests measuring significant enhancement of body and suppression of non-body cells’ responses in HBTs as time window to define high baseline activity varied over time…………………………….125
Figure 32. P-values of t-tests measuring significantly larger RMI values of body and smaller RMI values of non-body cells’ in HBTs vs. LBTs as time window to define high vs. low baseline activity varied over time…….….126
Figure 33. Body cells’ RMI values of high and low baseline trials across body stimulus signal levels…………………………………………….…127
Figure 34. Baseline dependent modulation of body and non-body cells’ responses to presentation of object images…………………………….…129
Figure 35. Baseline dependent modulation of body and non-body cells’ responses to presentation of object images…………………………….…130
Figure 36. Temporal dynamics of body and non-body cells’ RMI to presentation of object images in correct and wrong conditions for HBTs………………………………………………………………………131
Figure 37. Temporal dynamics of body and non-body cells’ RMI to presentation of object images in correct and wrong conditions for LBTs………………………………………………………………………132
Figure 38. Frequency distribution of adjusted RMI values for body and non-body cells in HBTs and LBTs………………………………………….…133
Figure 39. Rate-matched fano factor modulation index (FFMI) of body and non-body cells to presentation of body images in correct condition for HBTs vs. LBTs………………………………………………………………..…134
Figure 40. Rate-matched fano factor modulation (FFMI) of body and non-body cells to presentation of object images in correct conditions for HBTs and LBTs. …………………………………………………………………135
Figure 41a. Frequency distribution of normalized d’ modulation difference in LBTs vs. HBTs for body and non-body cells…………………………..136
Figure 41b. The impact of task specific attentional modulation on firing rate depends on cells’ category selectivity……………………………………..137
Figure 42. The impact of task specific attentional modulation on firing rate depends on cells’ category selectivity………………………………….…138
Figure 43. Comparison of RMI values of correct vs. wrong trials of LBTs and HBTs……………………………………………………………….…139
Figure 44. Comparison of rate modulation in body and non-body cells population across trials of body images with different baseline spike counts……………………………………………………………………..140
Figure 45. Comparison of rate modulation in body and non-body cells population across trials of object images with different baseline spike counts………………………………………………………………….….142
Figure 46. Baseline dependent correlation of neural activity and behavioral choice…………………………………………………………………..…143
Figure 47. Correlation between CP and cells’ body/object discrimination power………………………………………………………………………144
Figure 48. CP values of body cells plotted against the HBTs proportion in active task………………………………………………………………….145
Figure 49. RMI values of body cells plotted against the HBTs proportion in active task…………………………………………………………………146
Figure 50. Attentional modulation of baseline and evoked response in 30 low baseline cells…………………………………………………………147
Figure 51. Attentional modulation of baseline and evoked response in 30 low baseline cells…………………………………………………………148
Figure 52. Percent of HBT in active is plotted vs. percent of HBT in passive for 14 body and 16 non-body cells………………………………………..149
Figure 53. RMI of low baseline body and non-body cells in different stimulus and choice conditions……………………………………………150
Figure 54. RMI of low baseline body and non-body cells in different stimulus and choice conditions……………………………………………151
Figure 55. Percent of HBT is active vs. percent of HBT in passive………152
Figure 56. A combination of baseline firing rate and evoked response modulation in active compared with passive conditions affects monkeys’ performance…………………………………………………………….…153
Figure 57. Polar plots of IT cells activity show that baseline dependent differential response of IT cell subpopulations determines monkey’s choice…………………………………………………………………..…155
Appendix1: Stimulus set……………………………………………….……….158
Appendix2: List of abbreviations………………………………………………164
References………………………………………………………………………..166
Introduction
The crucial role of “visual object categorization” in everyday life
Our normal life relies on ability of visual object recognition or determining the identity of a seen object. We recognize different familiar or novel objects in everyday life. We do this with no or little effort, despite the fact that these objects may vary in form, color, illumination, view, size or texture from time to time. Based on both behavioral and neural data there are different levels of object recognition. When we see Einstein’s face, first we detect it as a “face” (supraordinate level), perceive as a “human face” (ordinate level) and then “Einstein’s face” (subordinate level). Spector and Kanwisher explored the sequence of processing steps in object recognition by asking human subjects to do three different tasks: object detection, categorization and identification. Accuracy and reaction time were similar for object detection and categorization showing that “as soon as you know it is there, you know what it is” (Spector and Kanwisher, 2005). On the other hand, lower accuracy and longer reaction time was observed for identification compared to categorization, introducing them as different steps of object recognition. Compatible with behavioral data firing patterns of single cells in inferior temporal cortex, a cortical area involved in object recognition, convey the information about categorization and identification with different latencies. Earliest part of the response (~120 ms after stimulus presentation) represents information about categorization while more detailed information about members of category started ~50 ms later (Sugase et al., 1999). Therefore, visual cortex processes information from global to fine in a hierarchical fashion. It has been suggested that categorization relies on the “presence or absence of features”, whereas identification is based on “configurational judgments”.
“Visual object categorization” or our ability to classify objects by giving meaning to our environment enables us to interact normally and efficiently with objects and events. There are some defined classes of objects in our mind. They usually share some major common properties in their appearance, while at the same time there are lots of differences among their members. For example, trees usually grow from the earth, they have roots, stem and usually green leaves. While they have similar properties, each of the species has a set of specific characteristics. But we call all of them trees, and also easily classify any new member as tree, even if we have not seen something like it before. This fascinating ability of categorization objects is vital for our survival. We know special traits for different object categories. We have learned how to treat and interact with any of them, depending on their characteristics. For example, classifying a rod-shaped moving object as “snake” makes us to run away as fast as possible. We perform this task easily and rapidly under very different conditions and even in noisy environment. Behavioral studies in human have shown that they can recognize animals in a cluttered picture which is presented only for 20ms with reaction times less than 400ms and 95% accuracy (Thorpe et al., 1996; Keysers et al., 2001). Monkeys showed even faster reaction times (Fabre-Thorpe et al., 1998). Monkeys could categorize food and trees with reaction times less than 250ms (Vogels, 1999a). Single cell studies in macaque inferior temporal (IT) cortex have revealed that category response latency is less than 100ms from stimulus onset (Vogels, 1999b; Kiani et al., 2005; Perrett et al., 1982).
Where in the brain is category information represented?
Neural mechanisms of and cortical areas involved in visual object categorization are among the hottest areas in field of cognitive neuroscience. Exploring the underlying mechanisms of visual categorization in the activity of single neurons of a special cortical area is based on what Santiago Ramon Cajal proposed by “Neuron Doctrine” over a century ago. He showed that nervous system is not one continuous web but a network of discrete cells. According to “Neuron Doctrine” individual neurons are the basic structural and functional units of the nervous system. This finding led to a new view of brain function called “Cellular Connectionism”. Based on this view, individual neurons are the signaling units of the brain; they are generally arranged in functional groups and connect to one another in a precise fashion and different behaviors are produced by different brain regions interconnected by specific neural pathways (Kandel, 2000).
Visual cortices are regions of the brain dedicated to the process of visual information. There is a “feed-forward flow of visual information” in these cortical areas. Visual information after reaching the eyes extends from the retina to the primary visual cortex (V1) and then the secondary visual cortex (V2). After V2, visual information goes through two different visual pathways:
1. Dorsa visual pathway or “what” pathway, involved in motion detection and visumotor tasks
2. Ventral visual pathway or “where” pathway, involved in object recognition
Understanding and recognition of shape of visual objects are completed in ventral visual pathway of the brain. Across the ventral visual pathway, there is a flow of visual information from the lower level visual areas (V1 & V2) into mid level (V4) and then to the high level visual area (IT) (Merigan & Maunsell, 1993). There is also a hierarchical organization even along the subareas of IT cortex. These intrinsic connections in the IT cortex were studied by Fujita & Fujita (1996). They showed that these connections were distributed in an anisotropic manner (fibers go through anteroposterior direction more than mediolateral direction) around the injection of the tracer showing the continuous feed-forward flow of visual information even in these subareas. Along with this feed-forward flow of visual information there is a hierarchical processing of the visual information. Reflected light from visual stimuli after entering the eyes is converted into electrical signals by photoreceptors and ganglion cells in the retina which respond optimally to contrast and small spots of light in their small receptive fields resulting in decomposition of visual stimuli into a pattern of small spots. Progressive convergence of input from retina and LGN (lateral geniculate nucleus) to the primary visual cortex (V1) leads to some feature abstraction. The outline of a visual image is decomposed into spots in retina and then recomposed into short line segments of various orientations by simple and complex cells in V1 cortex (Hubel & Wiesel, 1962). The visual pathway extends from V1 to V2. V2 neurons continue the analysis of contours begun by V1 neurons. Response of many V2 neurons to illusory lines just as real edges shows that the feature abstraction is in progress through the visual stream (Kandel, 2000). To clarify the progressive abstraction of visual information processing from V2 to downstream cortices, Kobatake & Tanaka (1994) defined an index based on the ratio of the maximum neural response to simple stimuli to the maximum neural response to all other stimuli in their image set (both simple and complex stimuli). The distribution of this ratio shifted from 1 toward 0 step by step from V2 to anterior IT. They showed that in macaque monkeys, the best stimulus of cells in V2 were just simple shapes, in V4 and posterior IT were both simple and complex features and the cells selective to complex features were intermingled in single penetrations with cells that responded maximally to some simple features. They also found that neurons of anterior IT were just selective to complex features. They suggested that local neuronal networks in V4 and posterior IT play an essential role in the abstraction of simple features into complex object features. These findings are consistent with “Feature Detection Theory”, one of the main theories in object recognition. According to this theory, the object perception proceeds by recognizing individual features, such as back, seat, arms and base of a chair, and assembling them into a coherent pattern, or chair.
The highest level and last purely visual processing of object recognition occurs at inferior temporal cortex (IT). Lesion studies in monkeys have shown that bilateral ablation or differentiation of IT cortex from upstream cortices would cause profound deficits in visual recognition, identification and categorization (Gross et al., 1972; Dean, 1976). The feed forward hierarchical processing of visual information reaches its most complete form in the IT cortex. Category selective cells have been reported only in “IT cortex” of monkeys and not in the lower cortical areas. Face cells in IT or single neurons that respond to faces more than any other categories were the first category neurons discovered by Charles Gross at the beginning of the 1970s (Gross et al., 1972). These cells respond to the whole combination of a set of features found in faces and are thought to be developed for a holistic processing of face as an especially important object. These face cells were defined by the firing rate of their responses which was significantly higher for face category compared to the others. Coding of categories has also been shown in other characteristics of neural response. Using information theory, Sugase et al. (1999) showed that there was category information (categorizing stimuli as monkey faces, human faces or shapes) in earliest part of IT neuronal responses. Coding of categories has also been reported in the onset latency of IT neuronal responses. Kiani et al. (2005) showed that many cells responded to human and non-primate animal faces with comparable magnitudes but responded significantly more quickly to human faces than to non-primate animal faces. They concluded that differences in onset latency may be used to increase the coding capacity. Representation of categories other than faces in IT was reported by Kiani et al. (2007). They introduced body category cells for the first time. Body cells responded to body images more than any other categories. Kiani et al also showed that different object categories including faces and bodies were represented not only in response patterns of single neurons, but also in neural population in monkeys’ anterior IT cortex. They showed that based on the response correlation of a population of IT neurons, objects can be categorized. Category information in neural population was also reported by Hung and colleagues (Hung et al, 2005). They used a biologically plausible, classifier-based readout technique to investigate the neural coding of selectivity at the IT population level. The activity of small neuronal populations (~100 randomly selected cells) over very short time intervals (as small as 12.5 milliseconds) contained unexpectedly accurate and robust information about both object ‘‘category’’ and ‘‘identity’’. This information generalized over a range of object positions and scales, even for novel objects. Functional imaging studies in human has also revealed cortical areas devoted to the processing of special categories which are socially and evolutionary important like fusiform face area (FFA), extrastriate body area (EBA) and parahippocampal place area (PPA) specialized for face, body and scene processing, respectively (Kanwisher and McDermott, 1997; Epstein and Kanwisher, 1998; Downing et al, 2001). Using fMRI in alert fixating macaque monkeys patches of maximally responsive to faces and bodies are found in visual cortex (Tsao et al, 2003). The accurate and rapid recognition of these well-familiar categories which are socially and evolutionary highly important, might be achieved by more developed representational mechanisms of these highly selective patches.
Most of the previous studies reported category selective cells are mainly located in the anterior parts of the IT cortex. Kiani et al. (2007) showed that category representation is present in more anterior parts of IT (anteroposterior position: 15 to 20 mm). TE is heavily connected with visual areas involved in memory (MTL or medial temporal lobe structures including perirhinal cortex, entorhinal cortex, hippocampus and parahippocampal gyrus) and decision making (prefrontal cortex or PFC).
There is some evidence showing that category information is present in neural populations which are either “selective or non-selective” to that category. Tsao et al. (2006) to measure how much information face-selective cells carry about face and other categories used a method based on the mean firing rate of population of IT neurons. They tested how accurate responses of these face selective neurons could predict the category of different objects. They showed that mean face categorization accuracy was 100%, and this value for other categories including bodies, fruits, technological gadgets, hands and grid scrambled patterns was above chance level. It is also shown that not only neural responses to the best categories, but also those to other categories, ranging from suboptimal to the worst carry category information (Kiani et al., 2007). These facts are consistent with what Haxby et al. (2001) reported on both distributed and overlapping category representation in human ventral temporal cortex. Using fMRI signal, they showed that distinct patterns exist for different categories. The distinctiveness of the response to a given category was not due simply to the regions that responded maximally to that category, because the category being viewed also could be identified on the basis of the pattern of response when those regions were excluded from the analysis. These findings show that there is a redundancy in category representation. What this redundancy is used for has remained to be answered. This overlapping representation might be important for processing of ambiguous visual information. In this situation, category representation might change from sparse coding to a more distributed one by using the available information in non-selective sites as well. On the other hand, the robustness of neurons with different category selectivity, to ambiguous images of preferred category members could be different due to distinct underlying representational mechanisms of preferred categories. Although some fMRI evidence shows that there is evolutionary evolved domain specificity in visual cortex for some socially important categories like faces or bodies, there are still shortcomings in comparing the representational mechanisms of these special categories with others at the level of single neurons.
There are also category representation in neurons of downstream cortices from IT like mediotemporal lobe (MTL) and prefrontal cortex (PFC). Kreiman et al. (2000) reported category-specific visual responses of single neurons in human MTL. Figure 9 shows an animal selective neuron. Wilson et al. (1993) showed that some neurons of PFC respond selectively to faces. MTL consisting of memory related cortices is important for object recognition but is not proposed as a cortical area directly affecting monkey’s decision. Albeit participation of perirhinal cortex as a component of MTL in semantic memory (Davies et al., 2004) makes it important for object recognition (Buckley et al., 2001) there is no evidence of its role as a cortical area directly involved in visual categorization. The main role of PFC is considered as matching of thoughts and actions according to the internal goals. Considering both category responses reported in single neurons of PFC during passively viewing of objects and its high engagement in doing active tasks, PFC has been proposed as a cortical area responsible for visual categorization. There is a debate over the role of category representation in IT and PFC to find out which of them is contributing to the behavior as the main cortical area involved in visual categorization. Since the presence of category representation in a cortical area per se does not indicate the role of that area in visual categorization, this issue should be investigated in neural responses of monkeys while they are doing visual categorization task. Different studies have shown that both IT and PFC are actively involved in visual categorization, but their respective roles are not known. Whether visual categorization is done by PFC neurons and this information is merely copied to the IT via direct interconnections between them or the opposite, remains to be answered. Comparing and contrasting the neural properties of different brain regions can make clear their respective contributions and so the neural circuitry underlying a given function. PFC is known to be involved in planning complex cognitive behaviors. Therefore, when a novel categorization task should be done, it could be solved by PFC. There is a possibility that for well-established categorization of well-familiar categories like faces or bodies, this role is transferred to IT as an upstream cortical area. The rapid, effortless and accurate recognition of these highly familiar objects supports this possibility. Freedman et al. (2003) claimed that category information is copied from PFC to IT by recording the activity of both PFC and IT neurons while monkeys were doing a cat/dog categorization task with different levels of morph between these two categories. They showed category response in both IT and PFC neurons while monkeys were doing the categorization task. Although they suggested that PFC played the main role in this categorization task, their data did not support it. They have reported the onset latency of neural response in IT (mean: 102.8 ms) and PFC (mean: 195.5 ms) to stimulus presentation, also latencies for neurons to begin to distinguish between categories in IT (mean: 126.7 ms) and PFC (mean: 171.1 ms). The onset of category response in IT is not only sooner than the same in PFC, but also sooner than onset latency of neural response in PFC neurons which rules out the possibility that this information is copied by PFC to IT via feedback connections. On the other hand, categorization of morphed cat/dog is a complicated kind of categorization task with novel stimuli (compared to well-familiar categories). Even if they had proved that this task was done by PFC, it would not have been surprising. Considering the presence of category information in IT before PFC in their data, according to the “principal of simplicity in biological systems”, it could be inferred that IT contributes in monkey’s choice.
The role of IT in visual perception and categorization and its relation with the behavior has been recently a matter of intense debate in cognitive neuroscience. There are some characteristics in the neural responses of IT cortex that are often taken as indirect evidence for their role in visual perception. First, they are selectively activated by complex visual stimuli like faces (Desimone et al., 1984); there is categorical representation of visual stimuli (kiani et al., 2007) and category information (Sugase et al., 1999) in single neurons and neural population of IT cortex. Second, they show considerable invariance in their stimulus selectivity with regard to variations of the position, size and luminance of the external stimuli (Tanaka, 1996). This helps generalization which is needed for categorization. Third, they are influenced by visual experience and training (Logothetis et al., 1995) and fourth they are selective to diagnostic features of categories (Sigala & Logothetis, 2002). To clarify the active role of IT in visual categorization, comparison of the response of the same neurons in passive (passively viewing) and active (categorization) tasks could be helpful. Koida and Komatsu (2006) tested the IT neurons response in a color categorization task, a color discrimination task and a simple fixation task. They showed that many IT neurons changed their activity depending upon the task. A majority of neurons showed stronger responses during the categorization task. Moreover, for the population of IT neurons as a whole, signals contributing to performing the categorization task were enhanced. These results imply that color-selective IT neurons’ activity is modulated in a task-dependent way. This question about the role of IT neurons in more complicated categorization tasks than color categorization has recently emerged as a central theme in neurophysiological studies of cognition and remained to be answered.
Understanding of how the human brain controls behavior is a crucial theme in cognitive neuroscience. Afraz et al. (2006) artificially activated small clusters of IT neurons by means of electrical microstimulation while the monkeys performed a face/ non- face categorization task. They showed that microstimulation of just face selective sites biased the monkeys’ decisions towards the face category. This causal relationship between the activity of face-selective neurons and face perception supports the possibility if IT role in visual categorization. This issue could be explored in more detail, by recording IT neural response while monkeys are doing a categorization task to compare these neural responses with monkey’s behavior on a trial by trial basis. “Prediction of the behavior from the neural response” was first explored in dorsal visual pathway by Britten et al. (1996). They studied the relationship between the discharge of MT neurons and behavioral choice. They found that trial-to-trial variability in neural signals was correlated with the choices the monkey made. They concluded that neuronal signals in MT are used by the monkey to determine the direction of stimulus motion. Correlation between the responses of disparity-selective neurons in IT and the behavioral discrimination of stereoscopic depth on a trial-to-trial basis was later shown by Uka et al. (2005). Mruczek & Sheinberg (2007) reported that the activity of IT neurons predicts whether or not a monkey will recognize and respond to a complex visual object. Correlation between the IT neural response and monkey’s choice while doing a categorization task is going to be explored in this project. “Comparing the neural and psychophysical sensitivity to stimulus change” as another method to explore the correlation between the neural response and behavior, was first tested by Britten et al. (1992) in MT cortex of macaques performing a direction discrimination task. They added different levels of ambiguity to the stimuli to make a spectrum of available visual information. Monkey’s behavioral performance in different levels of ambiguity (psychometric function) was compared to the neuronal performance (neurometric function) or the ability of neurons to discriminate between different stimuli. By showing the similarity between sensitivity of most MT neurons and the psychophysical sensitivity, they concluded that psychophysical decisions were likely to be based on neural signals. Liu & Jagadeesh (2008) used the same method of analysis to determine whether IT neurons can yield enough information to classify complex images. They recorded from IT neurons during the presentation of morphed photographic images in sessions in which monkeys classified images in a two-alternative forced-choice delayed match to sample (2AFC-DMS) task or in sessions in which they performed a fixation task. By reporting the sensitivity between neural and behavioral performance just when monkeys were doing the DMS task they concluded that IT contributes to the ability to discriminate and classify complex images. There were different shortcomings in their work which are followed. The sample stimuli were chosen from a sequence in which two colored original images were gradually morphed into each other. The choice stimuli were the original images. Morphing of two images into each other makes some features vague easier than the others. So monkeys could use them as some cues to do the task without any need to discriminate between complex images. On the other hand they compared neural data of both active and passive tasks with monkey’s behavior to show that the similarity between neural and behavioral performance existed just in active task. This comparison would have been possible if they had collected the active and passive data from the same neurons, while their two sets of neural data were collected in separate sessions and from separate neurons. Besides all these problems their task was a discrimination task in the form of DMS and not a categorization task.
The similarity of mechanisms used for visual categorization in human and monkey have been widely explored to see whether it is correct to use monkeys as a model to understand what is going on in human brain. Rosch and her colleagues (1976a) showed that human conceptual categories have a perceptual basis and categories actually reflect more than just linguistic constructs. Perceptual categories are determined by the high correlational structure of the real world, in which certain combinations of attributes are more probable than others. Bundles of such co-occurring attributes form the basis of a natural classification for objects. Cues with high frequency within a given category and low frequency in all other categories are valid category predictors (Logothetis, 1996). Categorization of novel stimuli in adults showed that subjects recognized the prototype pattern almost as quickly and as accurately as they recognize previously memorized patterns, even though they had never been directly exposed to it, suggesting that the prototype is a main constituent in the category’s memorial representation (Franks & Bransford, 1971, Strange et al., 1970). Like human adults, infants tend to abstract the central tendency of a category when exposed to a sufficient number of exemplars (Quinn & Eimas, 1986). Piaget (1969), after carefully observing the development of perception and cognition in children, concluded that these capacities are rooted in prelinguistic constructs that are only later enriched through the use of language. Nonhuman primates clearly are capable of making various categorizations at different abstraction levels (Lorenz, 1971; Davis, 1974). Monkeys, for instance, can learn to perform various types of complex classification tasks in the laboratory (Davis, 1974). They are obviously capable of making basic-level categorizations, but they can also easily learn to discriminate individual human or monkey faces (Rosenfeld & Van Hoesen, 1979; CJ Bruce, 1982) and novel artificial object classes, even generalizing learning across basic image transformations (Logothetis et a1., 1994, 1995). In order to see whether monkeys can categorize the visual stimuli in the same way as human does, Vogels (1999) trained rhesus monkeys in a visual categorization task designed in order to minimize rote learning of individual exemplars and to allow testing of transfer from old to novel exemplars of the category. The monkeys learned to distinguish complex color images of trees from other objects and generalized from a learned set of complex images to novel, physically dissimilar images. Considering the variability among tree images, categorization should be based on a combination of common features and perceptual learning instead of just exemplars and rote learning (Vogels, 1999).
There is also some evidence showing that not only the behavioral performance of human and monkey resembles each other, but also the underlying neural mechanism of such behavior are similar. Tsao et al. (2003) by using fMRI in alert fixating monkeys, found face and body patches similar in relative size and number to those in human. They concluded that humans and macaques share similar brain architecture for visual object processing. Kiani et al. (2007) showed the presence of an inherently categorical representation in naive monkeys from reconstruction a large part of our intuitive category structure in IT population responses. Similarity of category information representation in human and monkey IT cell population has been shown by comparing human fMRI and monkey single cell data (Kriegeskorte et al, 2008)
Anatomy of inferior temporal cortex
Neurons in the inferior temporal cortex (IT), an area crucially involved in visual object recognition and categorization, show the visual response properties, anatomical and chemical characteristics which are distinct from other cortical areas that feed visual inputs to the IT. Earlier physiological studies showed that IT neurons have large receptive fields covering the center and contralateral (often bilateral) visual fields, stimulus selectivity for images of complex objects or shapes, and translation invariance of the stimulus selectivity (Fujita, 2002). IT is generally considered a large and elongated region of cortex extending approximately from just anterior to the inferior occipital sulcus to a couple of millimeters posterior to the temporal pole (about 20 millimeters length), and from the fundus of the superior temporal sulcus (STS) to the fundus of the occipito-temporal sulcus. It is roughly coextensive with Brodmann areas 20 and 21 (Logothetis, 1996).
Based on the findings of lesion study, cytoarchitectonic properties, anatomical connections (both cortical and subcortical) and neural response properties including receptive field size and stimulus selectivity, IT cortex is considered as a modular structure with nine subareas:

در این سایت فقط تکه هایی از این مطلب با شماره بندی انتهای صفحه درج می شود که ممکن است هنگام انتقال از فایل ورد به داخل سایت کلمات به هم بریزد یا شکل ها درج نشود

شما می توانید تکه های دیگری از این مطلب را با جستجو در همین سایت بخوانید

ولی برای دانلود فایل اصلی با فرمت ورد حاوی تمامی قسمت ها با منابع کامل

اینجا کلیک کنید

1. Lower bank of STS, posterior part
2. Lower bank of STS, middle part
3. Lower bank of STS, anterior part
4. TEO, dorsal
5. TEO, ventral
6. TE, posterior & dorsal (TEpd)
7. TE, posterior & ventral (TEpv)
8. TE, anterior & dorsal (TEad)

دسته بندی : پایان نامه ارشد

دیدگاهتان را بنویسید