Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1298-0
M. Dujmović, Gaurav Malhotra, J. Bowers
In recent years, deep convolutional neural networks (DCNNs) have shown extraordinary success in object recognition tasks. However, they can also be fooled by adversarial images (stimuli designed to fool networks) that do not appear to fool humans. This has been taken as evidence that these models work quite differently than the human visual system. However, Zhou and Firestone (2019) carried out a study where they presented adversarial images which fool DCNNs to humans and found that, in many cases, humans chose the same label for these images as DCNNs. They take these findings to support the claim that human and machine vision is more similar than commonly claimed. Here we report two experiments that show that the level of agreement between human and DCNN classification is driven by how the experimenter chooses the adversarial images and how they choose the labels given to humans for classification. Based on how one chooses these variables, humans can show a span of agreement levels with DCNNs; from well below to well above levels expected by chance. Overall, our results do not support a view of large systematic overlap between human and computer vision.
{"title":"Humans cannot decipher adversarial images: Revisiting Zhou and Firestone (2019)","authors":"M. Dujmović, Gaurav Malhotra, J. Bowers","doi":"10.32470/ccn.2019.1298-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1298-0","url":null,"abstract":"In recent years, deep convolutional neural networks (DCNNs) have shown extraordinary success in object recognition tasks. However, they can also be fooled by adversarial images (stimuli designed to fool networks) that do not appear to fool humans. This has been taken as evidence that these models work quite differently than the human visual system. However, Zhou and Firestone (2019) carried out a study where they presented adversarial images which fool DCNNs to humans and found that, in many cases, humans chose the same label for these images as DCNNs. They take these findings to support the claim that human and machine vision is more similar than commonly claimed. Here we report two experiments that show that the level of agreement between human and DCNN classification is driven by how the experimenter chooses the adversarial images and how they choose the labels given to humans for classification. Based on how one chooses these variables, humans can show a span of agreement levels with DCNNs; from well below to well above levels expected by chance. Overall, our results do not support a view of large systematic overlap between human and computer vision.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130688477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1236-0
K. Frolichs, B. Kuper-Smith, J. Gläscher, Gabriela Rosenblau, C. Korn
Humans often describe other persons (and themselves) in terms of abstract character traits. When getting to know a new person, they need to update their estimates of the other person across many different character traits. It is unclear how this learning process unfolds and how the relationship between diverse character traits are represented in brain activity. Here, we first showed in three behavioral studies that humans combine reinforcement learning with their knowledge about the correlations between traits when learning about other persons’ character. Second, in two functional imaging studies the fine-grained similarities between character traits were represented in medial prefrontal cortex, in a region that has consistently been linked to thinking about other persons. Our findings thus suggest a possible learning mechanism for rather complex generalization across character traits according to their similarities, which seem to be related to the medial prefrontal cortex.
{"title":"Learning about Other Persons’ Character Traits Relies on Combining Reinforcement Learning with Representations of Trait Similarities","authors":"K. Frolichs, B. Kuper-Smith, J. Gläscher, Gabriela Rosenblau, C. Korn","doi":"10.32470/ccn.2019.1236-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1236-0","url":null,"abstract":"Humans often describe other persons (and themselves) in terms of abstract character traits. When getting to know a new person, they need to update their estimates of the other person across many different character traits. It is unclear how this learning process unfolds and how the relationship between diverse character traits are represented in brain activity. Here, we first showed in three behavioral studies that humans combine reinforcement learning with their knowledge about the correlations between traits when learning about other persons’ character. Second, in two functional imaging studies the fine-grained similarities between character traits were represented in medial prefrontal cortex, in a region that has consistently been linked to thinking about other persons. Our findings thus suggest a possible learning mechanism for rather complex generalization across character traits according to their similarities, which seem to be related to the medial prefrontal cortex.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130748313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1073-0
Nicholas Menghi, W. Penny
This paper describes a set of associative learning experiments in which the appropriate response depends on multiple relevant stimuli. We vary both the complexity of the stimulus-response mapping (task) and the temporal structure of the stimuli that are presented. We find that both of these manipulations affect the accuracy with which the task can be learnt, and that task complexity affects the proportion of subjects who correctly provide declarative knowledge of the underlying association. Computational modelling of subjects’ behaviour, based on Dynamic Logistic Regression models, allowed us to probe the strategies that subjects employed during learning. We found that the majority of subjects employed a configural learning strategy during the complex task and a mixed configural/rule-based strategy during the simpler task. Computational modelling also provided an entropybased index of strategy exploration with greater exploration observed during the complex task.
{"title":"Configural Learning depends on Task Complexity and Temporal Structure","authors":"Nicholas Menghi, W. Penny","doi":"10.32470/ccn.2019.1073-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1073-0","url":null,"abstract":"This paper describes a set of associative learning experiments in which the appropriate response depends on multiple relevant stimuli. We vary both the complexity of the stimulus-response mapping (task) and the temporal structure of the stimuli that are presented. We find that both of these manipulations affect the accuracy with which the task can be learnt, and that task complexity affects the proportion of subjects who correctly provide declarative knowledge of the underlying association. Computational modelling of subjects’ behaviour, based on Dynamic Logistic Regression models, allowed us to probe the strategies that subjects employed during learning. We found that the majority of subjects employed a configural learning strategy during the complex task and a mixed configural/rule-based strategy during the simpler task. Computational modelling also provided an entropybased index of strategy exploration with greater exploration observed during the complex task.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133381989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1422-0
Daniel N. Scott, M. Frank
Mechanistic models of cognitive phenomena often make use of neural networks, which allow researchers to examine relationships between neurobiology and the computations suspected to underlie cognition. These models typically make use of neural firing rates, as do analyses of in-vivo data, with the dimension of neural dynamics receiving special attention. Treating time-binned spiking activity as a sequence of binary vectors (spike-words) should prove complementary to rate-space analyses, and has been shown to provide links with statistical physics. We investigate the interaction between these two analyses using theory and simulations to show how signatures of rate-dynamics are found in spike-word distributions. We find that a global integration over the eigenvalues of linear dynamics local to attracting subspaces can modify spike-synchrony, and we quantify how this impacts informational and thermodynamic properties of these systems. The research outlined here will have implications for the interpretation of neural data, the use of population codes for tasks such as Bayesian inference, and for various resource rational models attempting to bridge the gap between computation and implementation.
{"title":"Rate-space attractors and low dimensional dynamics interact with spike-synchrony statistics in neural networks","authors":"Daniel N. Scott, M. Frank","doi":"10.32470/ccn.2019.1422-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1422-0","url":null,"abstract":"Mechanistic models of cognitive phenomena often make use of neural networks, which allow researchers to examine relationships between neurobiology and the computations suspected to underlie cognition. These models typically make use of neural firing rates, as do analyses of in-vivo data, with the dimension of neural dynamics receiving special attention. Treating time-binned spiking activity as a sequence of binary vectors (spike-words) should prove complementary to rate-space analyses, and has been shown to provide links with statistical physics. We investigate the interaction between these two analyses using theory and simulations to show how signatures of rate-dynamics are found in spike-word distributions. We find that a global integration over the eigenvalues of linear dynamics local to attracting subspaces can modify spike-synchrony, and we quantify how this impacts informational and thermodynamic properties of these systems. The research outlined here will have implications for the interpretation of neural data, the use of population codes for tasks such as Bayesian inference, and for various resource rational models attempting to bridge the gap between computation and implementation.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133581361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1173-0
Nicolas Perez-Nieves, Vincent C. H. Leung, P. Dragotti, Dan F. M. Goodman
It is very common in studies of the learning capabilities of spiking neural networks (SNNs) to use homogeneous neural and synaptic parameters (time constants, thresholds, etc.). Even in studies in which these parameters are distributed heterogeneously, the advantages or disadvantages of the heterogeneity have rarely been studied in depth. By contrast, in the brain, neurons and synapses are highly diverse, leading naturally to the hypothesis that this heterogeneity may be advantageous for learning. Starting from two state-of-the-art methods for training spiking neural networks (Nicola & Clopath, 2017; Shrestha & Orchard, 2018), we found that adding parameter heterogeneity reduced errors when the network had to learn more complex patterns, increased robustness to hyperparameter mistuning, and reduced the number of training iterations required. We propose that neural heterogeneity may be an important principle for brains to learn robustly in real world environments with highly complex structure, and where task-specific hyperparameter tuning may be impossible. Consequently, heterogeneity may also be a good candidate design principle for artificial neural networks, to reduce the need for expensive hyperparameter tuning as well as for reducing training time.
{"title":"Advantages of heterogeneity of parameters in spiking neural network training","authors":"Nicolas Perez-Nieves, Vincent C. H. Leung, P. Dragotti, Dan F. M. Goodman","doi":"10.32470/ccn.2019.1173-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1173-0","url":null,"abstract":"It is very common in studies of the learning capabilities of spiking neural networks (SNNs) to use homogeneous neural and synaptic parameters (time constants, thresholds, etc.). Even in studies in which these parameters are distributed heterogeneously, the advantages or disadvantages of the heterogeneity have rarely been studied in depth. By contrast, in the brain, neurons and synapses are highly diverse, leading naturally to the hypothesis that this heterogeneity may be advantageous for learning. Starting from two state-of-the-art methods for training spiking neural networks (Nicola & Clopath, 2017; Shrestha & Orchard, 2018), we found that adding parameter heterogeneity reduced errors when the network had to learn more complex patterns, increased robustness to hyperparameter mistuning, and reduced the number of training iterations required. We propose that neural heterogeneity may be an important principle for brains to learn robustly in real world environments with highly complex structure, and where task-specific hyperparameter tuning may be impossible. Consequently, heterogeneity may also be a good candidate design principle for artificial neural networks, to reduce the need for expensive hyperparameter tuning as well as for reducing training time.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133123675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1149-0
N. Seijdel, Nikos Tsakmakidis, E. Haan, S. Bohté, S. Scholte
Feedforward deep convolutional neural networks (DCNNs) are matching and even surpassing human performance on object recognition. This performance suggests that activation of a loose collection of image features could support the recognition of natural object categories, without dedicated systems to solve specific visual subtasks. Recent findings in humans however, suggest that while feedforward activity may suffice for sparse scenes with isolated objects, additional visual operations ('routines') that aid the recognition process (e.g. segmentation or grouping) are needed for more complex scenes. Linking human visual processing to performance of DCNNs with increasing depth, we here explored if, how, and when object information is differentiated from the backgrounds they appear on. To this end, we controlled the information in both objects and backgrounds, as well as the relationship between them by adding noise, manipulating background congruence and systematically occluding parts of the image. Results indicated less distinction between objectand background features for more shallow networks. For those networks, we observed a benefit of training on segmented objects (as compared to unsegmented objects). Overall, deeper networks trained on natural (unsegmented) scenes seem to perform implicit 'segmentation' of the objects from their background, possibly by improved selection of relevant features.
{"title":"Implicit Scene Segmentation in Deeper Convolutional Neural Networks","authors":"N. Seijdel, Nikos Tsakmakidis, E. Haan, S. Bohté, S. Scholte","doi":"10.32470/ccn.2019.1149-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1149-0","url":null,"abstract":"Feedforward deep convolutional neural networks (DCNNs) are matching and even surpassing human performance on object recognition. This performance suggests that activation of a loose collection of image features could support the recognition of natural object categories, without dedicated systems to solve specific visual subtasks. Recent findings in humans however, suggest that while feedforward activity may suffice for sparse scenes with isolated objects, additional visual operations ('routines') that aid the recognition process (e.g. segmentation or grouping) are needed for more complex scenes. Linking human visual processing to performance of DCNNs with increasing depth, we here explored if, how, and when object information is differentiated from the backgrounds they appear on. To this end, we controlled the information in both objects and backgrounds, as well as the relationship between them by adding noise, manipulating background congruence and systematically occluding parts of the image. Results indicated less distinction between objectand background features for more shallow networks. For those networks, we observed a benefit of training on segmented objects (as compared to unsegmented objects). Overall, deeper networks trained on natural (unsegmented) scenes seem to perform implicit 'segmentation' of the objects from their background, possibly by improved selection of relevant features.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"516 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133233126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1338-0
Angela Radulescu, Y. Niv, N. Daw
A growing literature has highlighted a role for selective attention in shaping representation learning of relevant task features, yet little is known about how humans learn what to attend to. Here we model the dynamics of selective attention as a memory-augmented particle filter. In a task where participants had to learn from trial and error which of nine features is more predictive of reward, we show that trial-by-trial attention to features measured with eye-tracking is better fit by the particle filter, compared to a reinforcement learning mechanism that had been proposed in the past. This is because inference based on a single particle captures the sparse allocation and rapid switching of attention better than incremental error-driven updates. However, because a single particle maintains insufficient information about past events to switch hypotheses as efficiently as do participants, we show that the data are best fit by the filter augmented with a memory buffer for recent observations. This proposal suggests a new role for memory in enabling tractable, resource-efficient approximations to normative inference.
{"title":"A particle filtering account of selective attention during learning","authors":"Angela Radulescu, Y. Niv, N. Daw","doi":"10.32470/ccn.2019.1338-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1338-0","url":null,"abstract":"A growing literature has highlighted a role for selective attention in shaping representation learning of relevant task features, yet little is known about how humans learn what to attend to. Here we model the dynamics of selective attention as a memory-augmented particle filter. In a task where participants had to learn from trial and error which of nine features is more predictive of reward, we show that trial-by-trial attention to features measured with eye-tracking is better fit by the particle filter, compared to a reinforcement learning mechanism that had been proposed in the past. This is because inference based on a single particle captures the sparse allocation and rapid switching of attention better than incremental error-driven updates. However, because a single particle maintains insufficient information about past events to switch hypotheses as efficiently as do participants, we show that the data are best fit by the filter augmented with a memory buffer for recent observations. This proposal suggests a new role for memory in enabling tractable, resource-efficient approximations to normative inference.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131688084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1293-0
M. Hajnal, Duy T. Tran, Michael C. Einstein, Gergő Orbán, P. Golshani, Pierre-Olivier Polack
{"title":"Neural signatures of coping with multiple tasks in mouse visual cortex","authors":"M. Hajnal, Duy T. Tran, Michael C. Einstein, Gergő Orbán, P. Golshani, Pierre-Olivier Polack","doi":"10.32470/ccn.2019.1293-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1293-0","url":null,"abstract":"","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115235721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1196-0
Libi Kliger, G. Yovel
According to the normalization framework the neural response of a single neuron to multiple stimuli is normalized by the response of its surrounding neurons. High-level visual cortex is composed of clusters of neurons that are selective to the same category. In an fMRI study, we show that the normalization model, together with the profile of category-selectivity of a given cortical area, can predict its response to multi-category stimuli. We measured the response to a face and a body (or a face and an object) presented alone or simultaneously and estimated the contribution of each category to the multicategory representation by fitting a linear model. Results show that the response to multi-category stimuli is a weighted mean of the response to each of its components. The coefficients were correlated with the selectivity profile of the cortical region. These findings suggest that the functional organization of category-selective cortex, i.e., neighboring patches of neurons, each selective to a single category, bias the response to certain categories, for which such clusters of neurons exist, and give them priority in the representation of cluttered visual scenes.
{"title":"Category-selectivity together with a Normalization Model Predicts the Response to Multi-category Stimuli along the Category-Selective Cortex","authors":"Libi Kliger, G. Yovel","doi":"10.32470/ccn.2019.1196-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1196-0","url":null,"abstract":"According to the normalization framework the neural response of a single neuron to multiple stimuli is normalized by the response of its surrounding neurons. High-level visual cortex is composed of clusters of neurons that are selective to the same category. In an fMRI study, we show that the normalization model, together with the profile of category-selectivity of a given cortical area, can predict its response to multi-category stimuli. We measured the response to a face and a body (or a face and an object) presented alone or simultaneously and estimated the contribution of each category to the multicategory representation by fitting a linear model. Results show that the response to multi-category stimuli is a weighted mean of the response to each of its components. The coefficients were correlated with the selectivity profile of the cortical region. These findings suggest that the functional organization of category-selective cortex, i.e., neighboring patches of neurons, each selective to a single category, bias the response to certain categories, for which such clusters of neurons exist, and give them priority in the representation of cluttered visual scenes.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115724239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.32470/ccn.2019.1150-0
J. Ritchie, Haemy Lee Masson, Stefania Bracci, H. O. D. Beeck
Representational similarity analysis (RSA) is increasingly part of the standard analytic toolkit in neuroimaging. Core to RSA is the measuring of neural dissimilarity between the response patterns for different conditions to construct neural representational dissimilarity matrices (RDMs). It has been proposed that noise normalizing these patterns, and using crossvalidated distances as a dissimilarity measure, is superior for characterizing the structure of neural RDMs. This assessment has been motivated by improvement in within-subject neural dissimilarity after noise normalization. However, between-subject reliability is more directly related to determining the amount of explainable variance, and the evaluation of observed effect sizes when they are correlated with behavioral or model RDMs. Across three datasets we did not find that noise normalization consistently boosts within-subject reliability, between-subject reliability or correlations with behavioral or model RDMs. Overall, our results provide equivocal support for the utility of noise normalization to RSA.
{"title":"The Unreliable Influence of Noise Normalization on the Reliability of Neural Dissimilarity","authors":"J. Ritchie, Haemy Lee Masson, Stefania Bracci, H. O. D. Beeck","doi":"10.32470/ccn.2019.1150-0","DOIUrl":"https://doi.org/10.32470/ccn.2019.1150-0","url":null,"abstract":"Representational similarity analysis (RSA) is increasingly part of the standard analytic toolkit in neuroimaging. Core to RSA is the measuring of neural dissimilarity between the response patterns for different conditions to construct neural representational dissimilarity matrices (RDMs). It has been proposed that noise normalizing these patterns, and using crossvalidated distances as a dissimilarity measure, is superior for characterizing the structure of neural RDMs. This assessment has been motivated by improvement in within-subject neural dissimilarity after noise normalization. However, between-subject reliability is more directly related to determining the amount of explainable variance, and the evaluation of observed effect sizes when they are correlated with behavioral or model RDMs. Across three datasets we did not find that noise normalization consistently boosts within-subject reliability, between-subject reliability or correlations with behavioral or model RDMs. Overall, our results provide equivocal support for the utility of noise normalization to RSA.","PeriodicalId":281121,"journal":{"name":"2019 Conference on Cognitive Computational Neuroscience","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115100552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：